MyArxiv
Computation and Language 89
☆ ModuleFormer: Learning Modular Large Language Models From Uncurated Data
Large Language Models (LLMs) have achieved remarkable results. But existing models are expensive to train and deploy, and it is also difficult to expand their knowledge beyond pre-training data without forgetting previous knowledge. This paper proposes a new neural network architecture, ModuleFormer, that leverages modularity to improve the efficiency and flexibility of large language models. ModuleFormer is based on the Sparse Mixture of Experts (SMoE). Unlike the previous SMoE-based modular language model [Gururangan et al., 2021], which requires domain-labeled data to learn domain-specific experts, ModuleFormer can induce modularity from uncurated data with its new load balancing and load concentration losses. ModuleFormer is a modular architecture that includes two different types of modules, new stick-breaking attention heads, and feedforward experts. Different modules are sparsely activated conditions on the input token during training and inference. In our experiment, we found that the modular architecture enables three important abilities for large pre-trained language models: 1) Efficiency, since ModuleFormer only activates a subset of its modules for each input token, thus it could achieve the same performance as dense LLMs with more than two times throughput; 2) Extendability, ModuleFormer is more immune to catastrophic forgetting than dense LLMs and can be easily extended with new modules to learn new knowledge that is not included in the training data; 3) Specialisation, finetuning ModuleFormer could specialize a subset of modules to the finetuning task, and the task-unrelated modules could be easily pruned for a lightweight deployment.
Transformers as Statisticians: Provable In-Context Learning with In-Context Algorithm Selection
Neural sequence models based on the transformer architecture have demonstrated remarkable \emph{in-context learning} (ICL) abilities, where they can perform new tasks when prompted with training and test examples, without any parameter update to the model. This work first provides a comprehensive statistical theory for transformers to perform ICL. Concretely, we show that transformers can implement a broad class of standard machine learning algorithms in context, such as least squares, ridge regression, Lasso, learning generalized linear models, and gradient descent on two-layer neural networks, with near-optimal predictive power on various in-context data distributions. Using an efficient implementation of in-context gradient descent as the underlying mechanism, our transformer constructions admit mild size bounds, and can be learned with polynomially many pretraining sequences. Building on these ``base'' ICL algorithms, intriguingly, we show that transformers can implement more complex ICL procedures involving \emph{in-context algorithm selection}, akin to what a statistician can do in real life -- A \emph{single} transformer can adaptively select different base ICL algorithms -- or even perform qualitatively different tasks -- on different input sequences, without any explicit prompting of the right algorithm or task. We both establish this in theory by explicit constructions, and also observe this phenomenon experimentally. In theory, we construct two general mechanisms for algorithm selection with concrete examples: pre-ICL testing, and post-ICL validation. As an example, we use the post-ICL validation mechanism to construct a transformer that can perform nearly Bayes-optimal ICL on a challenging task -- noisy linear models with mixed noise levels. Experimentally, we demonstrate the strong in-context algorithm selection capabilities of standard transformer architectures.
☆ On the Reliability of Watermarks for Large Language Models
Large language models (LLMs) are now deployed to everyday use and positioned to produce large quantities of text in the coming decade. Machine-generated text may displace human-written text on the internet and has the potential to be used for malicious purposes, such as spearphishing attacks and social media bots. Watermarking is a simple and effective strategy for mitigating such harms by enabling the detection and documentation of LLM-generated text. Yet, a crucial question remains: How reliable is watermarking in realistic settings in the wild? There, watermarked text might be mixed with other text sources, paraphrased by human writers or other language models, and used for applications in a broad number of domains, both social and technical. In this paper, we explore different detection schemes, quantify their power at detecting watermarks, and determine how much machine-generated text needs to be observed in each scenario to reliably detect the watermark. We especially highlight our human study, where we investigate the reliability of watermarking when faced with human paraphrasing. We compare watermark-based detection to other detection strategies, finding overall that watermarking is a reliable solution, especially because of its sample complexity - for all attacks we consider, the watermark evidence compounds the more examples are given, and the watermark is eventually detected.
comment: 14 pages in the main body. Code is available at https://github.com/jwkirchenbauer/lm-watermarking
☆ Revisiting Out-of-distribution Robustness in NLP: Benchmark, Analysis, and LLMs Evaluations
This paper reexamines the research on out-of-distribution (OOD) robustness in the field of NLP. We find that the distribution shift settings in previous studies commonly lack adequate challenges, hindering the accurate evaluation of OOD robustness. To address these issues, we propose a benchmark construction protocol that ensures clear differentiation and challenging distribution shifts. Then we introduce BOSS, a Benchmark suite for Out-of-distribution robustneSS evaluation covering 5 tasks and 20 datasets. Based on BOSS, we conduct a series of experiments on pre-trained language models for analysis and evaluation of OOD robustness. First, for vanilla fine-tuning, we examine the relationship between in-distribution (ID) and OOD performance. We identify three typical types that unveil the inner learning mechanism, which could potentially facilitate the forecasting of OOD robustness, correlating with the advancements on ID datasets. Then, we evaluate 5 classic methods on BOSS and find that, despite exhibiting some effectiveness in specific cases, they do not offer significant improvement compared to vanilla fine-tuning. Further, we evaluate 5 LLMs with various adaptation paradigms and find that when sufficient ID data is available, fine-tuning domain-specific models outperform LLMs on ID examples significantly. However, in the case of OOD instances, prioritizing LLMs with in-context learning yields better results. We identify that both fine-tuned small models and LLMs face challenges in effectively addressing downstream tasks. The code is public at \url{https://github.com/lifan-yuan/OOD_NLP}.
comment: Code is available at \url{https://github.com/lifan-yuan/OOD_NLP}
☆ The Two Word Test: A Semantic Benchmark for Large Language Models NeurIPS 2023
Large Language Models (LLMs) have shown remarkable abilities recently, including passing advanced professional exams and demanding benchmark tests. This performance has led many to suggest that they are close to achieving humanlike or 'true' understanding of language, and even Artificial General Intelligence (AGI). Here, we provide a new open-source benchmark that can assess semantic abilities of LLMs using two-word phrases using a task that can be performed relatively easily by humans without advanced training. Combining multiple words into a single concept is a fundamental aspect of human language and intelligence. The test requires meaningfulness judgments of 1768 noun-noun combinations that have been rated as meaningful (e.g., baby boy) or not meaningful (e.g., goat sky). by 150 human raters. We provide versions of the task that probe meaningfulness ratings on a 0-4 scale as well as binary judgments. We conducted a series of experiments using the TWT on GPT-4, GPT-3.5, and Bard, with both versions. Results demonstrated that, compared to humans, all models perform poorly at rating meaningfulness of these phrases. GPT-3.5 and Bard are also unable to make binary discriminations between sensible and nonsense phrases as making sense. GPT-4 makes a substantial improvement in binary discrimination of combinatorial phrases but is still significantly worse than human performance. The TWT can be used to understand the limitations and weaknesses of current LLMs, and potentially improve them. The test also reminds us that caution is warranted in attributing 'true understanding' or AGI to LLMs. TWT is available at: https://github.com/NickRiccardi/two-word-test
comment: 12 pages, 5 figures, 3 tables, submitted to NeurIPS 2023 Datasets and Benchmarks Track
☆ Language Models Get a Gender Makeover: Mitigating Gender Bias with Few-Shot Data Interventions ACL 2023
Societal biases present in pre-trained large language models are a critical issue as these models have been shown to propagate biases in countless downstream applications, rendering them unfair towards specific groups of people. Since large-scale retraining of these models from scratch is both time and compute-expensive, a variety of approaches have been previously proposed that de-bias a pre-trained model. While the majority of current state-of-the-art debiasing methods focus on changes to the training regime, in this paper, we propose data intervention strategies as a powerful yet simple technique to reduce gender bias in pre-trained models. Specifically, we empirically show that by fine-tuning a pre-trained model on only 10 de-biased (intervened) training examples, the tendency to favor any gender is significantly reduced. Since our proposed method only needs a few training examples, our few-shot debiasing approach is highly feasible and practical. Through extensive experimentation, we show that our debiasing technique performs better than competitive state-of-the-art baselines with minimal loss in language modeling ability.
comment: Accepted to ACL 2023 Main Conference
☆ Gender, names and other mysteries: Towards the ambiguous for gender-inclusive translation
The vast majority of work on gender in MT focuses on 'unambiguous' inputs, where gender markers in the source language are expected to be resolved in the output. Conversely, this paper explores the widespread case where the source sentence lacks explicit gender markers, but the target sentence contains them due to richer grammatical gender. We particularly focus on inputs containing person names. Investigating such sentence pairs casts a new light on research into MT gender bias and its mitigation. We find that many name-gender co-occurrences in MT data are not resolvable with 'unambiguous gender' in the source language, and that gender-ambiguous examples can make up a large proportion of training examples. From this, we discuss potential steps toward gender-inclusive translation which accepts the ambiguity in both gender and translation.
comment: GITT workshop at EAMT 2023
☆ ChatGPT is fun, but it is not funny! Humor is still challenging Large Language Models
Humor is a central aspect of human communication that has not been solved for artificial agents so far. Large language models (LLMs) are increasingly able to capture implicit and contextual information. Especially, OpenAI's ChatGPT recently gained immense public attention. The GPT3-based model almost seems to communicate on a human level and can even tell jokes. Humor is an essential component of human communication. But is ChatGPT really funny? We put ChatGPT's sense of humor to the test. In a series of exploratory experiments around jokes, i.e., generation, explanation, and detection, we seek to understand ChatGPT's capability to grasp and reproduce human humor. Since the model itself is not accessible, we applied prompt-based experiments. Our empirical evidence indicates that jokes are not hard-coded but mostly also not newly generated by the model. Over 90% of 1008 generated jokes were the same 25 Jokes. The system accurately explains valid jokes but also comes up with fictional explanations for invalid jokes. Joke-typical characteristics can mislead ChatGPT in the classification of jokes. ChatGPT has not solved computational humor yet but it can be a big leap toward "funny" machines.
☆ Multi-Task Training with In-Domain Language Models for Diagnostic Reasoning
Generative artificial intelligence (AI) is a promising direction for augmenting clinical diagnostic decision support and reducing diagnostic errors, a leading contributor to medical errors. To further the development of clinical AI systems, the Diagnostic Reasoning Benchmark (DR.BENCH) was introduced as a comprehensive generative AI framework, comprised of six tasks representing key components in clinical reasoning. We present a comparative analysis of in-domain versus out-of-domain language models as well as multi-task versus single task training with a focus on the problem summarization task in DR.BENCH (Gao et al., 2023). We demonstrate that a multi-task, clinically trained language model outperforms its general domain counterpart by a large margin, establishing a new state-of-the-art performance, with a ROUGE-L score of 28.55. This research underscores the value of domain-specific training for optimizing clinical diagnostic reasoning tasks.
comment: Accepted to 2023 ClinicalNLP Workshop
☆ Contrastive Bootstrapping for Label Refinement ACL 2023
Traditional text classification typically categorizes texts into pre-defined coarse-grained classes, from which the produced models cannot handle the real-world scenario where finer categories emerge periodically for accurate services. In this work, we investigate the setting where fine-grained classification is done only using the annotation of coarse-grained categories and the coarse-to-fine mapping. We propose a lightweight contrastive clustering-based bootstrapping method to iteratively refine the labels of passages. During clustering, it pulls away negative passage-prototype pairs under the guidance of the mapping from both global and local perspectives. Experiments on NYT and 20News show that our method outperforms the state-of-the-art methods by a large margin.
comment: ACL 2023
☆ Multimodal Learning Without Labeled Multimodal Data: Guarantees and Applications
In many machine learning systems that jointly learn from multiple modalities, a core research question is to understand the nature of multimodal interactions: the emergence of new task-relevant information during learning from both modalities that was not present in either alone. We study this challenge of interaction quantification in a semi-supervised setting with only labeled unimodal data and naturally co-occurring multimodal data (e.g., unlabeled images and captions, video and corresponding audio) but when labeling them is time-consuming. Using a precise information-theoretic definition of interactions, our key contributions are the derivations of lower and upper bounds to quantify the amount of multimodal interactions in this semi-supervised setting. We propose two lower bounds based on the amount of shared information between modalities and the disagreement between separately trained unimodal classifiers, and derive an upper bound through connections to approximate algorithms for min-entropy couplings. We validate these estimated bounds and show how they accurately track true interactions. Finally, two semi-supervised multimodal applications are explored based on these theoretical results: (1) analyzing the relationship between multimodal performance and estimated interactions, and (2) self-supervised learning that embraces disagreement between modalities beyond agreement as is typically done.
comment: Code available at: https://github.com/pliang279/PID
☆ Long-form analogies generated by chatGPT lack human-like psycholinguistic properties
Psycholinguistic analyses provide a means of evaluating large language model (LLM) output and making systematic comparisons to human-generated text. These methods can be used to characterize the psycholinguistic properties of LLM output and illustrate areas where LLMs fall short in comparison to human-generated text. In this work, we apply psycholinguistic methods to evaluate individual sentences from long-form analogies about biochemical concepts. We compare analogies generated by human subjects enrolled in introductory biochemistry courses to analogies generated by chatGPT. We perform a supervised classification analysis using 78 features extracted from Coh-metrix that analyze text cohesion, language, and readability (Graesser et. al., 2004). Results illustrate high performance for classifying student-generated and chatGPT-generated analogies. To evaluate which features contribute most to model performance, we use a hierarchical clustering approach. Results from this analysis illustrate several linguistic differences between the two sources.
comment: arxiv version of conference paper to appear at CogSci 2023 conference
PromptAttack: Probing Dialogue State Trackers with Adversarial Prompts ACL 2023
A key component of modern conversational systems is the Dialogue State Tracker (or DST), which models a user's goals and needs. Toward building more robust and reliable DSTs, we introduce a prompt-based learning approach to automatically generate effective adversarial examples to probe DST models. Two key characteristics of this approach are: (i) it only needs the output of the DST with no need for model parameters, and (ii) it can learn to generate natural language utterances that can target any DST. Through experiments over state-of-the-art DSTs, the proposed framework leads to the greatest reduction in accuracy and the best attack success rate while maintaining good fluency and a low perturbation ratio. We also show how much the generated adversarial examples can bolster a DST through adversarial training. These results indicate the strength of prompt-based attacks on DSTs and leave open avenues for continued refinement.
comment: To appear in Findings of ACL 2023
☆ Lenient Evaluation of Japanese Speech Recognition: Modeling Naturally Occurring Spelling Inconsistency ACL
Word error rate (WER) and character error rate (CER) are standard metrics in Speech Recognition (ASR), but one problem has always been alternative spellings: If one's system transcribes adviser whereas the ground truth has advisor, this will count as an error even though the two spellings really represent the same word. Japanese is notorious for ``lacking orthography'': most words can be spelled in multiple ways, presenting a problem for accurate ASR evaluation. In this paper we propose a new lenient evaluation metric as a more defensible CER measure for Japanese ASR. We create a lattice of plausible respellings of the reference transcription, using a combination of lexical resources, a Japanese text-processing system, and a neural machine translation model for reconstructing kanji from hiragana or katakana. In a manual evaluation, raters rated 95.4% of the proposed spelling variants as plausible. ASR results show that our method, which does not penalize the system for choosing a valid alternate spelling of a word, affords a 2.4%-3.1% absolute reduction in CER depending on the task.
comment: ACL Workshop on Computation and Written Language (CAWL) 2023
☆ Can current NLI systems handle German word order? Investigating language model performance on a new German challenge set of minimal pairs
Compared to English, German word order is freer and therefore poses additional challenges for natural language inference (NLI). We create WOGLI (Word Order in German Language Inference), the first adversarial NLI dataset for German word order that has the following properties: (i) each premise has an entailed and a non-entailed hypothesis; (ii) premise and hypotheses differ only in word order and necessary morphological changes to mark case and number. In particular, each premise andits two hypotheses contain exactly the same lemmata. Our adversarial examples require the model to use morphological markers in order to recognise or reject entailment. We show that current German autoencoding models fine-tuned on translated NLI data can struggle on this challenge set, reflecting the fact that translated NLI datasets will not mirror all necessary language phenomena in the target language. We also examine performance after data augmentation as well as on related word order phenomena derived from WOGLI. Our datasets are publically available at https://github.com/ireinig/wogli.
☆ Enhancing In-Context Learning with Answer Feedback for Multi-Span Question Answering NLPCC 2023
Whereas the recent emergence of large language models (LLMs) like ChatGPT has exhibited impressive general performance, it still has a large gap with fully-supervised models on specific tasks such as multi-span question answering. Previous researches found that in-context learning is an effective approach to exploiting LLM, by using a few task-related labeled data as demonstration examples to construct a few-shot prompt for answering new questions. A popular implementation is to concatenate a few questions and their correct answers through simple templates, informing LLM of the desired output. In this paper, we propose a novel way of employing labeled data such that it also informs LLM of some undesired output, by extending demonstration examples with feedback about answers predicted by an off-the-shelf model, e.g., correct, incorrect, or incomplete. Experiments on three multi-span question answering datasets as well as a keyphrase extraction dataset show that our new prompting strategy consistently improves LLM's in-context learning performance.
comment: 12 pages, submitted to NLPCC 2023
☆ Evaluation of ChatGPT on Biomedical Tasks: A Zero-Shot Comparison with Fine-Tuned Generative Transformers ACL 2023
ChatGPT is a large language model developed by OpenAI. Despite its impressive performance across various tasks, no prior work has investigated its capability in the biomedical domain yet. To this end, this paper aims to evaluate the performance of ChatGPT on various benchmark biomedical tasks, such as relation extraction, document classification, question answering, and summarization. To the best of our knowledge, this is the first work that conducts an extensive evaluation of ChatGPT in the biomedical domain. Interestingly, we find based on our evaluation that in biomedical datasets that have smaller training sets, zero-shot ChatGPT even outperforms the state-of-the-art fine-tuned generative transformer models, such as BioGPT and BioBART. This suggests that ChatGPT's pre-training on large text corpora makes it quite specialized even in the biomedical domain. Our findings demonstrate that ChatGPT has the potential to be a valuable tool for various tasks in the biomedical domain that lack large annotated data.
comment: Accepted by BioNLP@ACL 2023
☆ STEPS: A Benchmark for Order Reasoning in Sequential Tasks
Various human activities can be abstracted into a sequence of actions in natural text, i.e. cooking, repairing, manufacturing, etc. Such action sequences heavily depend on the executing order, while disorder in action sequences leads to failure of further task execution by robots or AI agents. Therefore, to verify the order reasoning capability of current neural models in sequential tasks, we propose a challenging benchmark , named STEPS. STEPS involves two subtask settings, focusing on determining the rationality of given next step in recipes and selecting the reasonable step from the multi-choice question, respectively. We describe the data construction and task formulations, and benchmark most of significant Large Language Models (LLMs). The experimental results demonstrate 1) The commonsense reasoning of action orders in sequential tasks are challenging to resolve via zero-shot prompting or few-shot in-context learning for LLMs; 2) Prompting method still significantly lags behind tuning-based method on STEPS.
comment: Work in Progress
☆ Zambezi Voice: A Multilingual Speech Corpus for Zambian Languages INTERSPEECH 2023
This work introduces Zambezi Voice, an open-source multilingual speech resource for Zambian languages. It contains two collections of datasets: unlabelled audio recordings of radio news and talk shows programs (160 hours) and labelled data (over 80 hours) consisting of read speech recorded from text sourced from publicly available literature books. The dataset is created for speech recognition but can be extended to multilingual speech processing research for both supervised and unsupervised learning approaches. To our knowledge, this is the first multilingual speech dataset created for Zambian languages. We exploit pretraining and cross-lingual transfer learning by finetuning the Wav2Vec2.0 large-scale multilingual pre-trained model to build end-to-end (E2E) speech recognition models for our baseline models. The dataset is released publicly under a Creative Commons BY-NC-ND 4.0 license and can be accessed through the project repository. See https://github.com/unza-speech-lab/zambezi-voice
comment: Accepted at INTERSPEECH 2023
☆ Examining Bias in Opinion Summarisation Through the Perspective of Opinion Diversity WASSA
Opinion summarisation is a task that aims to condense the information presented in the source documents while retaining the core message and opinions. A summary that only represents the majority opinions will leave the minority opinions unrepresented in the summary. In this paper, we use the stance towards a certain target as an opinion. We study bias in opinion summarisation from the perspective of opinion diversity, which measures whether the model generated summary can cover a diverse set of opinions. In addition, we examine opinion similarity, a measure of how closely related two opinions are in terms of their stance on a given topic, and its relationship with opinion diversity. Through the lens of stances towards a topic, we examine opinion diversity and similarity using three debatable topics under COVID-19. Experimental results on these topics revealed that a higher degree of similarity of opinions did not indicate good diversity or fairly cover the various opinions originally presented in the source documents. We found that BART and ChatGPT can better capture diverse opinions presented in the source documents.
comment: 9 pages, 3 figures, accepted at WASSA, ACL 2023
☆ Transfer Learning of Transformer-based Speech Recognition Models from Czech to Slovak
In this paper, we are comparing several methods of training the Slovak speech recognition models based on the Transformers architecture. Specifically, we are exploring the approach of transfer learning from the existing Czech pre-trained Wav2Vec 2.0 model into Slovak. We are demonstrating the benefits of the proposed approach on three Slovak datasets. Our Slovak models scored the best results when initializing the weights from the Czech model at the beginning of the pre-training phase. Our results show that the knowledge stored in the Cezch pre-trained model can be successfully reused to solve tasks in Slovak while outperforming even much larger public multilingual models.
comment: Accepted to TSD 2023
☆ M$^3$IT: A Large-Scale Dataset towards Multi-Modal Multilingual Instruction Tuning
Instruction tuning has significantly advanced large language models (LLMs) such as ChatGPT, enabling them to align with human instructions across diverse tasks. However, progress in open vision-language models (VLMs) has been limited due to the scarcity of high-quality instruction datasets. To tackle this challenge and promote research in the vision-language field, we introduce the Multi-Modal, Multilingual Instruction Tuning (M$^3$IT) dataset, designed to optimize VLM alignment with human instructions. Our M$^3$IT dataset comprises 40 carefully curated datasets, including 2.4 million instances and 400 manually written task instructions, reformatted into a vision-to-text structure. Key tasks are translated into 80 languages with an advanced translation system, ensuring broader accessibility. M$^3$IT surpasses previous datasets regarding task coverage, instruction number and instance scale. Moreover, we develop Ying-VLM, a VLM model trained on our M$^3$IT dataset, showcasing its potential to answer complex questions requiring world knowledge, generalize to unseen video tasks, and comprehend unseen instructions in Chinese. To encourage further research, we have open-sourced both the dataset and trained models.
comment: Dataset available at: https://huggingface.co/MMInstruction/M3IT
☆ Multilingual Clinical NER: Translation or Cross-lingual Transfer?
Natural language tasks like Named Entity Recognition (NER) in the clinical domain on non-English texts can be very time-consuming and expensive due to the lack of annotated data. Cross-lingual transfer (CLT) is a way to circumvent this issue thanks to the ability of multilingual large language models to be fine-tuned on a specific task in one language and to provide high accuracy for the same task in another language. However, other methods leveraging translation models can be used to perform NER without annotated data in the target language, by either translating the training set or test set. This paper compares cross-lingual transfer with these two alternative methods, to perform clinical NER in French and in German without any training data in those languages. To this end, we release MedNERF a medical NER test set extracted from French drug prescriptions and annotated with the same guidelines as an English dataset. Through extensive experiments on this dataset and on a German medical dataset (Frei and Kramer, 2021), we show that translation-based methods can achieve similar performance to CLT but require more care in their design. And while they can take advantage of monolingual clinical language models, those do not guarantee better results than large general-purpose multilingual models, whether with cross-lingual transfer or translation.
comment: 23 pages, Proceedings of the 5th Clinical Natural Language Processing Workshop
☆ Label Aware Speech Representation Learning For Language Identification
Speech representation learning approaches for non-semantic tasks such as language recognition have either explored supervised embedding extraction methods using a classifier model or self-supervised representation learning approaches using raw data. In this paper, we propose a novel framework of combining self-supervised representation learning with the language label information for the pre-training task. This framework, termed as Label Aware Speech Representation (LASR) learning, uses a triplet based objective function to incorporate language labels along with the self-supervised loss function. The speech representations are further fine-tuned for the downstream task. The language recognition experiments are performed on two public datasets - FLEURS and Dhwani. In these experiments, we illustrate that the proposed LASR framework improves over the state-of-the-art systems on language identification. We also report an analysis of the robustness of LASR approach to noisy/missing labels as well as its application to multi-lingual speech recognition tasks.
comment: Accepted at Interspeech 2023
☆ Arabic Dysarthric Speech Recognition Using Adversarial and Signal-Based Augmentation
Despite major advancements in Automatic Speech Recognition (ASR), the state-of-the-art ASR systems struggle to deal with impaired speech even with high-resource languages. In Arabic, this challenge gets amplified, with added complexities in collecting data from dysarthric speakers. In this paper, we aim to improve the performance of Arabic dysarthric automatic speech recognition through a multi-stage augmentation approach. To this effect, we first propose a signal-based approach to generate dysarthric Arabic speech from healthy Arabic speech by modifying its speed and tempo. We also propose a second stage Parallel Wave Generative (PWG) adversarial model that is trained on an English dysarthric dataset to capture language-independant dysarthric speech patterns and further augment the signal-adjusted speech samples. Furthermore, we propose a fine-tuning and text-correction strategies for Arabic Conformer at different dysarthric speech severity levels. Our fine-tuned Conformer achieved 18% Word Error Rate (WER) and 17.2% Character Error Rate (CER) on synthetically generated dysarthric speech from the Arabic commonvoice speech dataset. This shows significant WER improvement of 81.8% compared to the baseline model trained solely on healthy data. We perform further validation on real English dysarthric speech showing a WER improvement of 124% compared to the baseline trained only on healthy English LJSpeech dataset.
comment: Accepted to Interspeech 2023
☆ Youku-mPLUG: A 10 Million Large-scale Chinese Video-Language Dataset for Pre-training and Benchmarks
To promote the development of Vision-Language Pre-training (VLP) and multimodal Large Language Model (LLM) in the Chinese community, we firstly release the largest public Chinese high-quality video-language dataset named Youku-mPLUG, which is collected from Youku, a well-known Chinese video-sharing website, with strict criteria of safety, diversity, and quality. Youku-mPLUG contains 10 million Chinese video-text pairs filtered from 400 million raw videos across a wide range of 45 diverse categories for large-scale pre-training. In addition, to facilitate a comprehensive evaluation of video-language models, we carefully build the largest human-annotated Chinese benchmarks covering three popular video-language tasks of cross-modal retrieval, video captioning, and video category classification. Youku-mPLUG can enable researchers to conduct more in-depth multimodal research and develop better applications in the future. Furthermore, we release popular video-language pre-training models, ALPRO and mPLUG-2, and our proposed modularized decoder-only model mPLUG-video pre-trained on Youku-mPLUG. Experiments show that models pre-trained on Youku-mPLUG gain up to 23.1% improvement in video category classification. Besides, mPLUG-video achieves a new state-of-the-art result on these benchmarks with 80.5% top-1 accuracy in video category classification and 68.9 CIDEr score in video captioning, respectively. Finally, we scale up mPLUG-video based on the frozen Bloomz with only 1.7% trainable parameters as Chinese multimodal LLM, and demonstrate impressive instruction and video understanding ability. The zero-shot instruction understanding experiment indicates that pretraining with Youku-mPLUG can enhance the ability to comprehend overall and detailed visual semantics, recognize scene text, and leverage open-domain knowledge.
comment: Working in progress
☆ ConTextual Masked Auto-Encoder for Retrieval-based Dialogue Systems
Dialogue response selection aims to select an appropriate response from several candidates based on a given user and system utterance history. Recent studies have been improving the accuracy of dialogue response selection through post-training, mostly relying on naive masked language modeling methods. However, the recently developed generative methods have shown promising text representation capabilities in IR community, which could potentially lead to better dialogue semantics modeling. Thus, in this paper, we propose Dial-MAE (Dialogue Contextual Masking Auto-encoder), a straightforward yet effective post-training technique tailored for dialogue response selection. Dial-MAE uses an asymmetric encoder-decoder architecture that learns to better compress the semantics of the dialogue into dialogue-dense vectors. The process of Dial-MAE involves a deep encoder creating a dialogue embedding with the masked dialogue context, followed by a shallow decoder that uses this embedding along with the highly masked response to restore the original response. Our experiments have demonstrated that Dial-MAE is highly effective, achieving state-of-the-art performance on two commonly evaluated benchmarks.
GPT Self-Supervision for a Better Data Annotator
The task of annotating data into concise summaries poses a significant challenge across various domains, frequently requiring the allocation of significant time and specialized knowledge by human experts. Despite existing efforts to use large language models for annotation tasks, significant problems such as limited applicability to unlabeled data, the absence of self-supervised methods, and the lack of focus on complex structured data still persist. In this work, we propose a GPT self-supervision annotation method. This method embodies a generating-recovering paradigm that leverages the capabilities of one-shot learning capabilities in Generative Pretrained Transformer (GPT). The proposed approach comprises a one-shot tuning phase followed by a generation phase. In the one-shot tuning phase, we sample a data from the support set as part of the prompt for GPT to generate a textual summary, which is then used to recover the original data. The alignment score between the recovered and original data serves as a self-supervision navigator to refine the process. In the generation stage, the optimally selected one-shot sample serves as a template in the prompt and is applied to generating summaries from challenging datasets. The annotation performance is evaluated by tuning several human feedback reward networks and by calculating alignment scores between original and recovered data at both sentence and structure levels. Our self-supervised annotation method consistently achieves competitive scores, convincingly demonstrating its robust strength in various data-to-summary annotation tasks.
☆ World Models for Math Story Problems ACL
Solving math story problems is a complex task for students and NLP models alike, requiring them to understand the world as described in the story and reason over it to compute an answer. Recent years have seen impressive performance on automatically solving these problems with large pre-trained language models and innovative techniques to prompt them. However, it remains unclear if these models possess accurate representations of mathematical concepts. This leads to lack of interpretability and trustworthiness which impedes their usefulness in various applications. In this paper, we consolidate previous work on categorizing and representing math story problems and develop MathWorld, which is a graph-based semantic formalism specific for the domain of math story problems. With MathWorld, we can assign world models to math story problems which represent the situations and actions introduced in the text and their mathematical relationships. We combine math story problems from several existing datasets and annotate a corpus of 1,019 problems and 3,204 logical forms with MathWorld. Using this data, we demonstrate the following use cases of MathWorld: (1) prompting language models with synthetically generated question-answer pairs to probe their reasoning and world modeling abilities, and (2) generating new problems by using the world models as a design space.
comment: ACL Findings 2023
☆ Co-evolving Graph Reasoning Network for Emotion-Cause Pair Extraction ECML-PKDD 2023
Emotion-Cause Pair Extraction (ECPE) aims to extract all emotion clauses and their corresponding cause clauses from a document. Existing approaches tackle this task through multi-task learning (MTL) framework in which the two subtasks provide indicative clues for ECPE. However, the previous MTL framework considers only one round of multi-task reasoning and ignores the reverse feedbacks from ECPE to the subtasks. Besides, its multi-task reasoning only relies on semantics-level interactions, which cannot capture the explicit dependencies, and both the encoder sharing and multi-task hidden states concatenations can hardly capture the causalities. To solve these issues, we first put forward a new MTL framework based on Co-evolving Reasoning. It (1) models the bidirectional feedbacks between ECPE and its subtasks; (2) allows the three tasks to evolve together and prompt each other recurrently; (3) integrates prediction-level interactions to capture explicit dependencies. Then we propose a novel multi-task relational graph (MRG) to sufficiently exploit the causal relations. Finally, we propose a Co-evolving Graph Reasoning Network (CGR-Net) that implements our MTL framework and conducts Co-evolving Reasoning on MRG. Experimental results show that our model achieves new state-of-the-art performance, and further analysis confirms the advantages of our method.
comment: Accepted by ECML-PKDD 2023
☆ A Study on the Reliability of Automatic Dysarthric Speech Assessments
Automating dysarthria assessments offers the opportunity to develop effective, low-cost tools that address the current limitations of manual and subjective assessments. Nonetheless, it is unclear whether current approaches rely on dysarthria-related speech patterns or external factors. We aim toward obtaining a clearer understanding of dysarthria patterns. To this extent, we study the effects of noise in recordings, both through addition and reduction. We design and implement a new method for visualizing and comparing feature extractors and models, at a patient level, in a more interpretable way. We use the UA-Speech dataset with a speaker-based split of the dataset. Results reported in the literature appear to have been done irrespective of such split, leading to models that may be overconfident due to data-leakage. We hope that these results raise awareness in the research community regarding the requirements for establishing reliable automatic dysarthria assessment systems.
☆ Echoes from Alexandria: A Large Resource for Multilingual Book Summarization ACL 2023
In recent years, research in text summarization has mainly focused on the news domain, where texts are typically short and have strong layout features. The task of full-book summarization presents additional challenges which are hard to tackle with current resources, due to their limited size and availability in English only. To overcome these limitations, we present "Echoes from Alexandria", or in shortened form, "Echoes", a large resource for multilingual book summarization. Echoes features three novel datasets: i) Echo-Wiki, for multilingual book summarization, ii) Echo-XSum, for extremely-compressive multilingual book summarization, and iii) Echo-FairySum, for extractive book summarization. To the best of our knowledge, Echoes, with its thousands of books and summaries, is the largest resource, and the first to be multilingual, featuring 5 languages and 25 language pairs. In addition to Echoes, we also introduce a new extractive-then-abstractive baseline, and, supported by our experimental results and manual analysis of the summaries generated, we argue that this baseline is more suitable for book summarization than purely-abstractive approaches. We release our resource and software at https://github.com/Babelscape/echoes-from-alexandria in the hope of fostering innovative research in multilingual book summarization.
comment: 9 pages, long paper at ACL 2023
☆ IUTEAM1 at MEDIQA-Chat 2023: Is simple fine tuning effective for multilayer summarization of clinical conversations?
Clinical conversation summarization has become an important application of Natural language Processing. In this work, we intend to analyze summarization model ensembling approaches, that can be utilized to improve the overall accuracy of the generated medical report called chart note. The work starts with a single summarization model creating the baseline. Then leads to an ensemble of summarization models trained on a separate section of the chart note. This leads to the final approach of passing the generated results to another summarization model in a multi-layer/stage fashion for better coherency of the generated text. Our results indicate that although an ensemble of models specialized in each section produces better results, the multi-layer/stage approach does not improve accuracy. The code for the above paper is available at https://github.com/dhananjay-srivastava/MEDIQA-Chat-2023-iuteam1.git
comment: preprint
☆ Cross-Genre Argument Mining: Can Language Models Automatically Fill in Missing Discourse Markers?
Available corpora for Argument Mining differ along several axes, and one of the key differences is the presence (or absence) of discourse markers to signal argumentative content. Exploring effective ways to use discourse markers has received wide attention in various discourse parsing tasks, from which it is well-known that discourse markers are strong indicators of discourse relations. To improve the robustness of Argument Mining systems across different genres, we propose to automatically augment a given text with discourse markers such that all relations are explicitly signaled. Our analysis unveils that popular language models taken out-of-the-box fail on this task; however, when fine-tuned on a new heterogeneous dataset that we construct (including synthetic and real examples), they perform considerably better. We demonstrate the impact of our approach on an Argument Mining downstream task, evaluated on different corpora, showing that language models can be trained to automatically fill in discourse markers across different corpora, improving the performance of a downstream model in some, but not all, cases. Our proposed approach can further be employed as an assistive tool for better discourse understanding.
☆ Personality testing of GPT-3: Limited temporal reliability, but highlighted social desirability of GPT-3's personality instruments results
To assess the potential applications and limitations of chatbot GPT-3 Davinci-003, this study explored the temporal reliability of personality questionnaires applied to the chatbot and its personality profile. Psychological questionnaires were administered to the chatbot on two separate occasions, followed by a comparison of the responses to human normative data. The findings revealed varying levels of agreement in the chatbot's responses over time, with some scales displaying excellent while others demonstrated poor agreement. Overall, Davinci-003 displayed a socially desirable and pro-social personality profile, particularly in the domain of communion. However, the underlying basis of the chatbot's responses, whether driven by conscious self-reflection or predetermined algorithms, remains uncertain.
comment: 18 pages, 1 table
☆ Allophant: Cross-lingual Phoneme Recognition with Articulatory Attributes INTERSPEECH 2023
This paper proposes Allophant, a multilingual phoneme recognizer. It requires only a phoneme inventory for cross-lingual transfer to a target language, allowing for low-resource recognition. The architecture combines a compositional phone embedding approach with individually supervised phonetic attribute classifiers in a multi-task architecture. We also introduce Allophoible, an extension of the PHOIBLE database. When combined with a distance based mapping approach for grapheme-to-phoneme outputs, it allows us to train on PHOIBLE inventories directly. By training and evaluating on 34 languages, we found that the addition of multi-task learning improves the model's capability of being applied to unseen phonemes and phoneme inventories. On supervised languages we achieve phoneme error rate improvements of 11 percentage points (pp.) compared to a baseline without multi-task learning. Evaluation of zero-shot transfer on 84 languages yielded a decrease in PER of 2.63 pp. over the baseline.
comment: 5 pages, 2 figures, 2 tables, accepted to INTERSPEECH 2023
☆ Phrase Retrieval for Open-Domain Conversational Question Answering with Conversational Dependency Modeling via Contrastive Learning ACL 2023
Open-Domain Conversational Question Answering (ODConvQA) aims at answering questions through a multi-turn conversation based on a retriever-reader pipeline, which retrieves passages and then predicts answers with them. However, such a pipeline approach not only makes the reader vulnerable to the errors propagated from the retriever, but also demands additional effort to develop both the retriever and the reader, which further makes it slower since they are not runnable in parallel. In this work, we propose a method to directly predict answers with a phrase retrieval scheme for a sequence of words, reducing the conventional two distinct subtasks into a single one. Also, for the first time, we study its capability for ODConvQA tasks. However, simply adopting it is largely problematic, due to the dependencies between previous and current turns in a conversation. To address this problem, we further introduce a novel contrastive learning strategy, making sure to reflect previous turns when retrieving the phrase for the current context, by maximizing representational similarities of consecutive turns in a conversation while minimizing irrelevant conversational contexts. We validate our model on two ODConvQA datasets, whose experimental results show that it substantially outperforms the relevant baselines with the retriever-reader. Code is available at: https://github.com/starsuzi/PRO-ConvQA.
comment: Findings of ACL 2023
☆ Analysis of the Fed's communication by using textual entailment model of Zero-Shot classification
In this study, we analyze documents published by central banks using text mining techniques and propose a method to evaluate the policy tone of central banks. Since the monetary policies of major central banks have a broad impact on financial market trends, the pricing of risky assets, and the real economy, market participants are attempting to more accurately capture changes in the outlook for central banks' future monetary policies. Since the published documents are also an important tool for the central bank to communicate with the market, they are meticulously elaborated on grammatical syntax and wording, and investors are urged to read more accurately about the central bank's policy stance. Sentiment analysis on central bank documents has long been carried out, but it has been difficult to interpret the meaning of the documents accurately and to explicitly capture even the intentional change in nuance. This study attempts to evaluate the implication of the zero-shot text classification method for an unknown economic environment using the same model. We compare the tone of the statements, minutes, press conference transcripts of FOMC meetings, and the Fed officials' (chair, vice chair, and Governors) speeches. In addition, the minutes of the FOMC meetings were subjected to a phase analysis of changes in each policy stance since 1971.
comment: 6 pages, 4 figures, 2 Tables
☆ Multi-microphone Automatic Speech Segmentation in Meetings Based on Circular Harmonics Features ISCA
Speaker diarization is the task of answering Who spoke and when? in an audio stream. Pipeline systems rely on speech segmentation to extract speakers' segments and achieve robust speaker diarization. This paper proposes a common framework to solve three segmentation tasks in the distant speech scenario: Voice Activity Detection (VAD), Overlapped Speech Detection (OSD), and Speaker Change Detection (SCD). In the literature, a few studies investigate the multi-microphone distant speech scenario. In this work, we propose a new set of spatial features based on direction-of-arrival estimations in the circular harmonic domain (CH-DOA). These spatial features are extracted from multi-microphone audio data and combined with standard acoustic features. Experiments on the AMI meeting corpus show that CH-DOA can improve the segmentation while being robust in the case of deactivated microphones.
comment: Interspeech 2023, international Speech Communication Association (ISCA), Aug 2023, Dublin, Ireland
☆ Transfer Learning from Pre-trained Language Models Improves End-to-End Speech Summarization
End-to-end speech summarization (E2E SSum) directly summarizes input speech into easy-to-read short sentences with a single model. This approach is promising because it, in contrast to the conventional cascade approach, can utilize full acoustical information and mitigate to the propagation of transcription errors. However, due to the high cost of collecting speech-summary pairs, an E2E SSum model tends to suffer from training data scarcity and output unnatural sentences. To overcome this drawback, we propose for the first time to integrate a pre-trained language model (LM), which is highly capable of generating natural sentences, into the E2E SSum decoder via transfer learning. In addition, to reduce the gap between the independently pre-trained encoder and decoder, we also propose to transfer the baseline E2E SSum encoder instead of the commonly used automatic speech recognition encoder. Experimental results show that the proposed model outperforms baseline and data augmented models.
comment: Accepted by Interspeech 2023
☆ Effective Neural Topic Modeling with Embedding Clustering Regularization ICML 2023
Topic models have been prevalent for decades with various applications. However, existing topic models commonly suffer from the notorious topic collapsing: discovered topics semantically collapse towards each other, leading to highly repetitive topics, insufficient topic discovery, and damaged model interpretability. In this paper, we propose a new neural topic model, Embedding Clustering Regularization Topic Model (ECRTM). Besides the existing reconstruction error, we propose a novel Embedding Clustering Regularization (ECR), which forces each topic embedding to be the center of a separately aggregated word embedding cluster in the semantic space. This enables each produced topic to contain distinct word semantics, which alleviates topic collapsing. Regularized by ECR, our ECRTM generates diverse and coherent topics together with high-quality topic distributions of documents. Extensive experiments on benchmark datasets demonstrate that ECRTM effectively addresses the topic collapsing issue and consistently surpasses state-of-the-art baselines in terms of topic quality, topic distributions of documents, and downstream classification tasks.
comment: Accepted to ICML 2023 conference
☆ Leveraging Knowledge Graph Embeddings to Enhance Contextual Representations for Relation Extraction
Relation extraction task is a crucial and challenging aspect of Natural Language Processing. Several methods have surfaced as of late, exhibiting notable performance in addressing the task; however, most of these approaches rely on vast amounts of data from large-scale knowledge graphs or language models pretrained on voluminous corpora. In this paper, we hone in on the effective utilization of solely the knowledge supplied by a corpus to create a high-performing model. Our objective is to showcase that by leveraging the hierarchical structure and relational distribution of entities within a corpus without introducing external knowledge, a relation extraction model can achieve significantly enhanced performance. We therefore proposed a relation extraction approach based on the incorporation of pretrained knowledge graph embeddings at the corpus scale into the sentence-level contextual representation. We conducted a series of experiments which revealed promising and very interesting results for our proposed approach.The obtained results demonstrated an outperformance of our method compared to context-based relation extraction models.
comment: 15 pages, 1 figures, The 17th International Conference on Document Analysis and Recognition
☆ An ASR-Based Tutor for Learning to Read: How to Optimize Feedback to First Graders SP
The interest in employing automatic speech recognition (ASR) in applications for reading practice has been growing in recent years. In a previous study, we presented an ASR-based Dutch reading tutor application that was developed to provide instantaneous feedback to first-graders learning to read. We saw that ASR has potential at this stage of the reading process, as the results suggested that pupils made progress in reading accuracy and fluency by using the software. In the current study, we used children's speech from an existing corpus (JASMIN) to develop two new ASR systems, and compared the results to those of the previous study. We analyze correct/incorrect classification of the ASR systems using human transcripts at word level, by means of evaluation measures such as Cohen's Kappa, Matthews Correlation Coefficient (MCC), precision, recall and F-measures. We observe improvements for the newly developed ASR systems regarding the agreement with human-based judgment and correct rejection (CR). The accuracy of the ASR systems varies for different reading tasks and word types. Our results suggest that, in the current configuration, it is difficult to classify isolated words. We discuss these results, possible ways to improve our systems and avenues for future research.
comment: Published (double-blind peer-reviewed) on SPECOM 2021
☆ A New Dataset and Empirical Study for Sentence Simplification in Chinese ACL2023
Sentence Simplification is a valuable technique that can benefit language learners and children a lot. However, current research focuses more on English sentence simplification. The development of Chinese sentence simplification is relatively slow due to the lack of data. To alleviate this limitation, this paper introduces CSS, a new dataset for assessing sentence simplification in Chinese. We collect manual simplifications from human annotators and perform data analysis to show the difference between English and Chinese sentence simplifications. Furthermore, we test several unsupervised and zero/few-shot learning methods on CSS and analyze the automatic evaluation and human evaluation results. In the end, we explore whether Large Language Models can serve as high-quality Chinese sentence simplification systems by evaluating them on CSS.
comment: Accepted by ACL2023 main conference
☆ Knowing-how & Knowing-that: A New Task for Machine Reading Comprehension of User Manuals
The machine reading comprehension (MRC) of user manuals has huge potential in customer service. However,current methods have trouble answering complex questions. Therefore, we introduce the Knowing-how & Knowing-that task that requires the model to answer factoid-style, procedure-style, and inconsistent questions about user manuals. We resolve this task by jointly representing the steps and facts in a graph (TARA), which supports a unified inference of various questions. Towards a systematical benchmarking study, we design a heuristic method to automatically parse user manuals into TARAs and build an annotated dataset to test the model's ability in answering real-world questions. Empirical results demonstrate that representing user manuals as TARAs is a desired solution for the MRC of user manuals. An in-depth investigation of TARA further sheds light on the issues and broader impacts of future representations of user manuals. We hope our work can move the MRC of user manuals to a more complex and realistic stage.
☆ Benchmarking Foundation Models with Language-Model-as-an-Examiner
Numerous benchmarks have been established to assess the performance of foundation models on open-ended question answering, which serves as a comprehensive test of a model's ability to understand and generate language in a manner similar to humans. Most of these works focus on proposing new datasets, however, we see two main issues within previous benchmarking pipelines, namely testing leakage and evaluation automation. In this paper, we propose a novel benchmarking framework, Language-Model-as-an-Examiner, where the LM serves as a knowledgeable examiner that formulates questions based on its knowledge and evaluates responses in a reference-free manner. Our framework allows for effortless extensibility as various LMs can be adopted as the examiner, and the questions can be constantly updated given more diverse trigger topics. For a more comprehensive and equitable evaluation, we devise three strategies: (1) We instruct the LM examiner to generate questions across a multitude of domains to probe for a broad acquisition, and raise follow-up questions to engage in a more in-depth assessment. (2) Upon evaluation, the examiner combines both scoring and ranking measurements, providing a reliable result as it aligns closely with human annotations. (3) We additionally propose a decentralized Peer-examination method to address the biases in a single examiner. Our data and benchmarking results are available at: https://lmexam.com.
comment: 23 pages, 8 figures
☆ When to Read Documents or QA History: On Unified and Selective Open-domain QA ACL 2023
This paper studies the problem of open-domain question answering, with the aim of answering a diverse range of questions leveraging knowledge resources. Two types of sources, QA-pair and document corpora, have been actively leveraged with the following complementary strength. The former is highly precise when the paraphrase of given question $q$ was seen and answered during training, often posed as a retrieval problem, while the latter generalizes better for unseen questions. A natural follow-up is thus leveraging both models, while a naive pipelining or integration approaches have failed to bring additional gains over either model alone. Our distinction is interpreting the problem as calibration, which estimates the confidence of predicted answers as an indicator to decide when to use a document or QA-pair corpus. The effectiveness of our method was validated on widely adopted benchmarks such as Natural Questions and TriviaQA.
comment: Findings of ACL 2023 camera ready
☆ From the One, Judge of the Whole: Typed Entailment Graph Construction with Predicate Generation ACL 2023
Entailment Graphs (EGs) have been constructed based on extracted corpora as a strong and explainable form to indicate context-independent entailment relations in natural languages. However, EGs built by previous methods often suffer from the severe sparsity issues, due to limited corpora available and the long-tail phenomenon of predicate distributions. In this paper, we propose a multi-stage method, Typed Predicate-Entailment Graph Generator (TP-EGG), to tackle this problem. Given several seed predicates, TP-EGG builds the graphs by generating new predicates and detecting entailment relations among them. The generative nature of TP-EGG helps us leverage the recent advances from large pretrained language models (PLMs), while avoiding the reliance on carefully prepared corpora. Experiments on benchmark datasets show that TP-EGG can generate high-quality and scale-controllable entailment graphs, achieving significant in-domain improvement over state-of-the-art EGs and boosting the performance of down-stream inference tasks.
comment: 9 pages, 3 figures, accepted to ACL 2023
☆ Increasing Diversity While Maintaining Accuracy: Text Data Generation with Large Language Models and Human Interventions ACL 2023
Large language models (LLMs) can be used to generate text data for training and evaluating other models. However, creating high-quality datasets with LLMs can be challenging. In this work, we explore human-AI partnerships to facilitate high diversity and accuracy in LLM-based text data generation. We first examine two approaches to diversify text generation: 1) logit suppression, which minimizes the generation of languages that have already been frequently generated, and 2) temperature sampling, which flattens the token sampling probability. We found that diversification approaches can increase data diversity but often at the cost of data accuracy (i.e., text and labels being appropriate for the target domain). To address this issue, we examined two human interventions, 1) label replacement (LR), correcting misaligned labels, and 2) out-of-scope filtering (OOSF), removing instances that are out of the user's domain of interest or to which no considered label applies. With oracle studies, we found that LR increases the absolute accuracy of models trained with diversified datasets by 14.4%. Moreover, we found that some models trained with data generated with LR interventions outperformed LLM-based few-shot classification. In contrast, OOSF was not effective in increasing model accuracy, implying the need for future work in human-in-the-loop text data generation.
comment: Accepted as a long paper at ACL 2023
☆ Knowledge-Augmented Language Model Prompting for Zero-Shot Knowledge Graph Question Answering
Large Language Models (LLMs) are capable of performing zero-shot closed-book question answering tasks, based on their internal knowledge stored in parameters during pre-training. However, such internalized knowledge might be insufficient and incorrect, which could lead LLMs to generate factually wrong answers. Furthermore, fine-tuning LLMs to update their knowledge is expensive. To this end, we propose to augment the knowledge directly in the input of LLMs. Specifically, we first retrieve the relevant facts to the input question from the knowledge graph based on semantic similarities between the question and its associated facts. After that, we prepend the retrieved facts to the input question in the form of the prompt, which is then forwarded to LLMs to generate the answer. Our framework, Knowledge-Augmented language model PromptING (KAPING), requires no model training, thus completely zero-shot. We validate the performance of our KAPING framework on the knowledge graph question answering task, that aims to answer the user's question based on facts over a knowledge graph, on which ours outperforms relevant zero-shot baselines by up to 48% in average, across multiple LLMs of various sizes.
☆ Multimodal Fusion Interactions: A Study of Human and Automatic Quantification
Multimodal fusion of multiple heterogeneous and interconnected signals is a fundamental challenge in almost all multimodal problems and applications. In order to perform multimodal fusion, we need to understand the types of interactions that modalities can exhibit: how each modality individually provides information useful for a task and how this information changes in the presence of other modalities. In this paper, we perform a comparative study of how human annotators can be leveraged to annotate two categorizations of multimodal interactions: (1) partial labels, where different randomly assigned annotators annotate the label given the first, second, and both modalities, and (2) counterfactual labels, where the same annotator is tasked to annotate the label given the first modality before giving them the second modality and asking them to explicitly reason about how their answer changes, before proposing an alternative taxonomy based on (3) information decomposition, where annotators annotate the degrees of redundancy: the extent to which modalities individually and together give the same predictions on the task, uniqueness: the extent to which one modality enables a task prediction that the other does not, and synergy: the extent to which only both modalities enable one to make a prediction about the task that one would not otherwise make using either modality individually. Through extensive experiments and annotations, we highlight several opportunities and limitations of each approach and propose a method to automatically convert annotations of partial and counterfactual labels to information decomposition, yielding an accurate and efficient method for quantifying interactions in multimodal datasets.
☆ Unbalanced Optimal Transport for Unbalanced Word Alignment ACL 2023
Monolingual word alignment is crucial to model semantic interactions between sentences. In particular, null alignment, a phenomenon in which words have no corresponding counterparts, is pervasive and critical in handling semantically divergent sentences. Identification of null alignment is useful on its own to reason about the semantic similarity of sentences by indicating there exists information inequality. To achieve unbalanced word alignment that values both alignment and null alignment, this study shows that the family of optimal transport (OT), i.e., balanced, partial, and unbalanced OT, are natural and powerful approaches even without tailor-made techniques. Our extensive experiments covering unsupervised and supervised settings indicate that our generic OT-based alignment methods are competitive against the state-of-the-arts specially designed for word alignment, remarkably on challenging datasets with high null alignment frequencies.
comment: Accepted for the Annual Meeting of the Association for Computational Linguistics (ACL 2023)
☆ Gotta: Generative Few-shot Question Answering by Prompt-based Cloze Data Augmentation
Few-shot question answering (QA) aims at precisely discovering answers to a set of questions from context passages while only a few training samples are available. Although existing studies have made some progress and can usually achieve proper results, they suffer from understanding deep semantics for reasoning out the questions. In this paper, we develop Gotta, a Generative prOmpT-based daTa Augmentation framework to mitigate the challenge above. Inspired by the human reasoning process, we propose to integrate the cloze task to enhance few-shot QA learning. Following the recent success of prompt-tuning, we present the cloze task in the same format as the main QA task, allowing the model to learn both tasks seamlessly together to fully take advantage of the power of prompt-tuning. Extensive experiments on widely used benchmarks demonstrate that Gotta consistently outperforms competitive baselines, validating the effectiveness of our proposed prompt-tuning-based cloze task, which not only fine-tunes language models but also learns to guide reasoning in QA tasks. Further analysis shows that the prompt-based loss incorporates the auxiliary task better than the multi-task loss, highlighting the strength of prompt-tuning on the few-shot QA task.
☆ XSemPLR: Cross-Lingual Semantic Parsing in Multiple Natural Languages and Meaning Representations ACL 2023
Cross-Lingual Semantic Parsing (CLSP) aims to translate queries in multiple natural languages (NLs) into meaning representations (MRs) such as SQL, lambda calculus, and logic forms. However, existing CLSP models are separately proposed and evaluated on datasets of limited tasks and applications, impeding a comprehensive and unified evaluation of CLSP on a diverse range of NLs and MRs. To this end, we present XSemPLR, a unified benchmark for cross-lingual semantic parsing featured with 22 natural languages and 8 meaning representations by examining and selecting 9 existing datasets to cover 5 tasks and 164 domains. We use XSemPLR to conduct a comprehensive benchmark study on a wide range of multilingual language models including encoder-based models (mBERT, XLM-R), encoder-decoder models (mBART, mT5), and decoder-based models (Codex, BLOOM). We design 6 experiment settings covering various lingual combinations (monolingual, multilingual, cross-lingual) and numbers of learning samples (full dataset, few-shot, and zero-shot). Our experiments show that encoder-decoder models (mT5) achieve the highest performance compared with other popular models, and multilingual training can further improve the average performance. Notably, multilingual large language models (e.g., BLOOM) are still inadequate to perform CLSP tasks. We also find that the performance gap between monolingual training and cross-lingual transfer learning is still significant for multilingual models, though it can be mitigated by cross-lingual few-shot training. Our dataset and code are available at https://github.com/psunlpgroup/XSemPLR.
comment: ACL 2023
☆ Text-only Domain Adaptation using Unified Speech-Text Representation in Transducer
Domain adaptation using text-only corpus is challenging in end-to-end(E2E) speech recognition. Adaptation by synthesizing audio from text through TTS is resource-consuming. We present a method to learn Unified Speech-Text Representation in Conformer Transducer(USTR-CT) to enable fast domain adaptation using the text-only corpus. Different from the previous textogram method, an extra text encoder is introduced in our work to learn text representation and is removed during inference, so there is no modification for online deployment. To improve the efficiency of adaptation, single-step and multi-step adaptations are also explored. The experiments on adapting LibriSpeech to SPGISpeech show the proposed method reduces the word error rate(WER) by relatively 44% on the target domain, which is better than those of TTS method and textogram method. Also, it is shown the proposed method can be combined with internal language model estimation(ILME) to further improve the performance.
comment: Submitted to Interspeech 2023
♻ ☆ PALR: Personalization Aware LLMs for Recommendation
Large language models (LLMs) have recently received significant attention for their exceptional capabilities. Despite extensive efforts in developing general-purpose LLMs that can be utilized in various natural language processing (NLP) tasks, there has been less research exploring their potential in recommender systems. In this paper, we propose a novel framework, named PALR, which aiming to combine user history behaviors (such as clicks, purchases, ratings, etc.) with LLMs to generate user preferred items. Specifically, we first use user/item interactions as guidance for candidate retrieval. Then we adopt a LLM-based ranking model to generate recommended items. Unlike existing approaches that typically adopt general-purpose LLMs for zero/few-shot recommendation testing or training on small-sized language models (with less than 1 billion parameters), which cannot fully elicit LLMs' reasoning abilities and leverage rich item side parametric knowledge, we fine-tune a 7 billion parameters LLM for the ranking purpose. This model takes retrieval candidates in natural language format as input, with instruction which explicitly asking to select results from input candidates during inference. Our experimental results demonstrate that our solution outperforms state-of-the-art models on various sequential recommendation tasks.
♻ ☆ Language Models can Solve Computer Tasks
Agents capable of carrying out general tasks on a computer can improve efficiency and productivity by automating repetitive tasks and assisting in complex problem-solving. Ideally, such agents should be able to solve new computer tasks presented to them through natural language commands. However, previous approaches to this problem require large amounts of expert demonstrations and task-specific reward functions, both of which are impractical for new tasks. In this work, we show that a pre-trained large language model (LLM) agent can execute computer tasks guided by natural language using a simple prompting scheme where the agent Recursively Criticizes and Improves its output (RCI). The RCI approach significantly outperforms existing LLM methods for automating computer tasks and surpasses supervised learning (SL) and reinforcement learning (RL) approaches on the MiniWoB++ benchmark. We compare multiple LLMs and find that RCI with the InstructGPT-3+RLHF LLM is state-of-the-art on MiniWoB++, using only a handful of demonstrations per task rather than tens of thousands, and without a task-specific reward function. Furthermore, we demonstrate RCI prompting's effectiveness in enhancing LLMs' reasoning abilities on a suite of natural language reasoning tasks, outperforming chain of thought (CoT) prompting. We find that RCI combined with CoT performs better than either separately. Our code can be found here: https://github.com/posgnu/rci-agent.
♻ ☆ ChatDB: Augmenting LLMs with Databases as Their Symbolic Memory
Large language models (LLMs) with memory are computationally universal. However, mainstream LLMs are not taking full advantage of memory, and the designs are heavily influenced by biological brains. Due to their approximate nature and proneness to the accumulation of errors, conventional neural memory mechanisms cannot support LLMs to simulate complex reasoning. In this paper, we seek inspiration from modern computer architectures to augment LLMs with symbolic memory for complex multi-hop reasoning. Such a symbolic memory framework is instantiated as an LLM and a set of SQL databases, where the LLM generates SQL instructions to manipulate the SQL databases. We validate the effectiveness of the proposed memory framework on a synthetic dataset requiring complex reasoning. The project website is available at https://chatdatabase.github.io/ .
♻ ☆ Z-Code++: A Pre-trained Language Model Optimized for Abstractive Summarization ACL 2023
This paper presents Z-Code++, a new pre-trained language model optimized for abstractive text summarization. The model extends the state of the art encoder-decoder model using three techniques. First, we use a two-phase pre-training process to improve model's performance on low-resource summarization tasks. The model is first pre-trained using text corpora for language understanding, and then is continually pre-trained on summarization corpora for grounded text generation. Second, we replace self-attention layers in the encoder with disentangled attention layers, where each word is represented using two vectors that encode its content and position, respectively. Third, we use fusion-in-encoder, a simple yet effective method of encoding long sequences in a hierarchical manner. Z-Code++ creates new state of the art on 9 out of 13 text summarization tasks across 5 languages. Our model is parameter-efficient in that it outperforms the 600x larger PaLM-540B on XSum, and the finetuned 200x larger GPT3-175B on SAMSum. In zero-shot and few-shot settings, our model substantially outperforms the competing models.
comment: 16 pages, 3 figures. Accepted as long paper in main conference of ACL 2023
♻ ☆ Easily Accessible Text-to-Image Generation Amplifies Demographic Stereotypes at Large Scale
Machine learning models that convert user-written text descriptions into images are now widely available online and used by millions of users to generate millions of images a day. We investigate the potential for these models to amplify dangerous and complex stereotypes. We find a broad range of ordinary prompts produce stereotypes, including prompts simply mentioning traits, descriptors, occupations, or objects. For example, we find cases of prompting for basic traits or social roles resulting in images reinforcing whiteness as ideal, prompting for occupations resulting in amplification of racial and gender disparities, and prompting for objects resulting in reification of American norms. Stereotypes are present regardless of whether prompts explicitly mention identity and demographic language or avoid such language. Moreover, stereotypes persist despite mitigation strategies; neither user attempts to counter stereotypes by requesting images with specific counter-stereotypes nor institutional attempts to add system ``guardrails'' have prevented the perpetuation of stereotypes. Our analysis justifies concerns regarding the impacts of today's models, presenting striking exemplars, and connecting these findings with deep insights into harms drawn from social scientific and humanist disciplines. This work contributes to the effort to shed light on the uniquely complex biases in language-vision models and demonstrates the ways that the mass deployment of text-to-image generation models results in mass dissemination of stereotypes and resulting harms.
comment: FAccT 2023 paper. The published version is available at 10.1145/3593013.3594095
♻ ☆ SpokenWOZ: A Large-Scale Speech-Text Dataset for Spoken Task-Oriented Dialogue in Multiple Domains
Task-oriented dialogue (TOD) models have made significant progress in recent years. However, previous studies primarily focus on datasets written by annotators, which has resulted in a gap between academic research and real-world spoken conversation scenarios. While several small-scale spoken TOD datasets are proposed to address robustness issues such as ASR errors, they ignore the unique challenges in spoken conversation. To tackle the limitations, we introduce SpokenWOZ, a large-scale speech-text dataset for spoken TOD, containing 8 domains, 203k turns, 5.7k dialogues and 249 hours of audios from human-to-human spoken conversations. SpokenWOZ further incorporates common spoken characteristics such as word-by-word processing and reasoning in spoken language. Based on these characteristics, we present cross-turn slot and reasoning slot detection as new challenges. We conduct experiments on various baselines, including text-modal models, newly proposed dual-modal models, and LLMs, e.g., ChatGPT. The results show that the current models still have substantial room for improvement in spoken conversation, where the most advanced dialogue state tracker only achieves 25.65% in joint goal accuracy and the SOTA end-to-end model only correctly completes the user request in 52.1% of dialogues. The dataset, code, and leaderboard are available: https://spokenwoz.github.io/SpokenWOZ-github.io/.
♻ ☆ Extrapolative Controlled Sequence Generation via Iterative Refinement ICML 2023
We study the problem of extrapolative controlled generation, i.e., generating sequences with attribute values beyond the range seen in training. This task is of significant importance in automated design, especially drug discovery, where the goal is to design novel proteins that are \textit{better} (e.g., more stable) than existing sequences. Thus, by definition, the target sequences and their attribute values are out of the training distribution, posing challenges to existing methods that aim to directly generate the target sequence. Instead, in this work, we propose Iterative Controlled Extrapolation (ICE) which iteratively makes local edits to a sequence to enable extrapolation. We train the model on synthetically generated sequence pairs that demonstrate small improvement in the attribute value. Results on one natural language task (sentiment analysis) and two protein engineering tasks (ACE2 stability and AAV fitness) show that ICE considerably outperforms state-of-the-art approaches despite its simplicity. Our code and models are available at: https://github.com/vishakhpk/iter-extrapolation.
comment: ICML 2023 - Camera Ready Version
♻ ☆ Handling the Alignment for Wake Word Detection: A Comparison Between Alignment-Based, Alignment-Free and Hybrid Approaches
Wake word detection exists in most intelligent homes and portable devices. It offers these devices the ability to "wake up" when summoned at a low cost of power and computing. This paper focuses on understanding alignment's role in developing a wake-word system that answers a generic phrase. We discuss three approaches. The first is alignment-based, where the model is trained with frame-wise cross-entropy. The second is alignment-free, where the model is trained with CTC. The third, proposed by us, is a hybrid solution in which the model is trained with a small set of aligned data and then tuned with a sizeable unaligned dataset. We compare the three approaches and evaluate the impact of the different aligned-to-unaligned ratios for hybrid training. Our results show that the alignment-free system performs better than the alignment-based for the target operating point, and with a small fraction of the data (20%), we can train a model that complies with our initial constraints.
comment: Accepted to Interspeech 2023
♻ ☆ Multi-Party Chat: Conversational Agents in Group Settings with Humans and Models
Current dialogue research primarily studies pairwise (two-party) conversations, and does not address the everyday setting where more than two speakers converse together. In this work, we both collect and evaluate multi-party conversations to study this more general case. We use the LIGHT environment to construct grounded conversations, where each participant has an assigned character to role-play. We thus evaluate the ability of language models to act as one or more characters in such conversations. Models require two skills that pairwise-trained models appear to lack: (1) being able to decide when to talk; (2) producing coherent utterances grounded on multiple characters. We compare models trained on our new dataset to existing pairwise-trained dialogue models, as well as large language models with few-shot prompting. We find that our new dataset, MultiLIGHT, which we will publicly release, can help bring significant improvements in the group setting.
♻ ☆ ChatGPT an ENFJ, Bard an ISTJ: Empirical Study on Personalities of Large Language Models
Large Language Models (LLMs) have made remarkable advancements in the field of artificial intelligence, significantly reshaping the human-computer interaction. We not only focus on the performance of LLMs, but also explore their features from a psychological perspective, acknowledging the importance of understanding their behavioral characteristics. Our study examines the behavioral patterns displayed by LLMs by employing trait theory, a psychological framework. We first focus on evaluating the consistency of personality types exhibited by ChatGPT. Furthermore, experiments include cross-lingual effects on seven additional languages, and the investigation of six other LLMs. Moreover, the study investigates whether ChatGPT can exhibit personality changes in response to instructions or contextual cues. The findings show that ChatGPT consistently maintains its ENFJ personality regardless of instructions or contexts. By shedding light on the personalization of LLMs, we anticipate that our study will serve as a catalyst for further research in this field.
comment: Added robustness analysis against fine-tuning (results of text-davinci-003); Added results of ChatGLM; Added limitations
♻ ☆ Uncovering and Categorizing Social Biases in Text-to-SQL
Content Warning: This work contains examples that potentially implicate stereotypes, associations, and other harms that could be offensive to individuals in certain social groups.} Large pre-trained language models are acknowledged to carry social biases towards different demographics, which can further amplify existing stereotypes in our society and cause even more harm. Text-to-SQL is an important task, models of which are mainly adopted by administrative industries, where unfair decisions may lead to catastrophic consequences. However, existing Text-to-SQL models are trained on clean, neutral datasets, such as Spider and WikiSQL. This, to some extent, cover up social bias in models under ideal conditions, which nevertheless may emerge in real application scenarios. In this work, we aim to uncover and categorize social biases in Text-to-SQL models. We summarize the categories of social biases that may occur in structured data for Text-to-SQL models. We build test benchmarks and reveal that models with similar task accuracy can contain social biases at very different rates. We show how to take advantage of our methodology to uncover and assess social biases in the downstream Text-to-SQL task. We will release our code and data.
♻ ☆ Causal interventions expose implicit situation models for commonsense language understanding ACL
Accounts of human language processing have long appealed to implicit ``situation models'' that enrich comprehension with relevant but unstated world knowledge. Here, we apply causal intervention techniques to recent transformer models to analyze performance on the Winograd Schema Challenge (WSC), where a single context cue shifts interpretation of an ambiguous pronoun. We identify a relatively small circuit of attention heads that are responsible for propagating information from the context word that guides which of the candidate noun phrases the pronoun ultimately attends to. We then compare how this circuit behaves in a closely matched ``syntactic'' control where the situation model is not strictly necessary. These analyses suggest distinct pathways through which implicit situation models are constructed to guide pronoun resolution.
comment: Findings of ACL
♻ ☆ Exploring Anisotropy and Outliers in Multilingual Language Models for Cross-Lingual Semantic Sentence Similarity ACL
Previous work has shown that the representations output by contextual language models are more anisotropic than static type embeddings, and typically display outlier dimensions. This seems to be true for both monolingual and multilingual models, although much less work has been done on the multilingual context. Why these outliers occur and how they affect the representations is still an active area of research. We investigate outlier dimensions and their relationship to anisotropy in multiple pre-trained multilingual language models. We focus on cross-lingual semantic similarity tasks, as these are natural tasks for evaluating multilingual representations. Specifically, we examine sentence representations. Sentence transformers which are fine-tuned on parallel resources (that are not always available) perform better on this task, and we show that their representations are more isotropic. However, we aim to improve multilingual representations in general. We investigate how much of the performance difference can be made up by only transforming the embedding space without fine-tuning, and visualise the resulting spaces. We test different operations: Removing individual outlier dimensions, cluster-based isotropy enhancement, and ZCA whitening. We publish our code for reproducibility.
comment: To appear in ACL Findings 2023. Fixed a citation in this version
♻ ☆ A Context-Sensitive Word Embedding Approach for The Detection of Troll Tweets
In this study, we aimed to address the growing concern of trolling behavior on social media by developing and evaluating a set of model architectures for the automatic detection of troll tweets. Utilizing deep learning techniques and pre-trained word embedding methods such as BERT, ELMo, and GloVe, we evaluated the performance of each architecture using metrics such as classification accuracy, F1 score, AUC, and precision. Our results indicate that BERT and ELMo embedding methods performed better than the GloVe method, likely due to their ability to provide contextualized word embeddings that better capture the nuances and subtleties of language use in online social media. Additionally, we found that CNN and GRU encoders performed similarly in terms of F1 score and AUC, suggesting their effectiveness in extracting relevant information from input text. The best-performing method was found to be an ELMo-based architecture that employed a GRU classifier, with an AUC score of 0.929. This research highlights the importance of utilizing contextualized word embeddings and appropriate encoder methods in the task of troll tweet detection, which can assist social-based systems in improving their performance in identifying and addressing trolling behavior on their platforms.
♻ ☆ Improving Cancer Hallmark Classification with BERT-based Deep Learning Approach
This paper presents a novel approach to accurately classify the hallmarks of cancer, which is a crucial task in cancer research. Our proposed method utilizes the Bidirectional Encoder Representations from Transformers (BERT) architecture, which has shown exceptional performance in various downstream applications. By applying transfer learning, we fine-tuned the pre-trained BERT model on a small corpus of biomedical text documents related to cancer. The outcomes of our experimental investigations demonstrate that our approach attains a noteworthy accuracy of 94.45%, surpassing almost all prior findings with a substantial increase of at least 8.04% as reported in the literature. These findings highlight the effectiveness of our proposed model in accurately classifying and comprehending text documents for cancer research, thus contributing significantly to the field. As cancer remains one of the top ten leading causes of death globally, our approach holds great promise in advancing cancer research and improving patient outcomes.
♻ ☆ HowkGPT: Investigating the Detection of ChatGPT-generated University Student Homework through Context-Aware Perplexity Analysis
As the use of Large Language Models (LLMs) in text generation tasks proliferates, concerns arise over their potential to compromise academic integrity. The education sector currently tussles with distinguishing student-authored homework assignments from AI-generated ones. This paper addresses the challenge by introducing HowkGPT, designed to identify homework assignments generated by AI. HowkGPT is built upon a dataset of academic assignments and accompanying metadata [17] and employs a pretrained LLM to compute perplexity scores for student-authored and ChatGPT-generated responses. These scores then assist in establishing a threshold for discerning the origin of a submitted assignment. Given the specificity and contextual nature of academic work, HowkGPT further refines its analysis by defining category-specific thresholds derived from the metadata, enhancing the precision of the detection. This study emphasizes the critical need for effective strategies to uphold academic integrity amidst the growing influence of LLMs and provides an approach to ensuring fair and accurate grading in educational institutions.
♻ ☆ Using Bottleneck Adapters to Identify Cancer in Clinical Notes under Low-Resource Constraints
Processing information locked within clinical health records is a challenging task that remains an active area of research in biomedical NLP. In this work, we evaluate a broad set of machine learning techniques ranging from simple RNNs to specialised transformers such as BioBERT on a dataset containing clinical notes along with a set of annotations indicating whether a sample is cancer-related or not. Furthermore, we specifically employ efficient fine-tuning methods from NLP, namely, bottleneck adapters and prompt tuning, to adapt the models to our specialised task. Our evaluations suggest that fine-tuning a frozen BERT model pre-trained on natural language and with bottleneck adapters outperforms all other strategies, including full fine-tuning of the specialised BioBERT model. Based on our findings, we suggest that using bottleneck adapters in low-resource situations with limited access to labelled data or processing capacity could be a viable strategy in biomedical text mining. The code used in the experiments are going to be made available at https://github.com/omidrohanian/bottleneck-adapters.
♻ ☆ Saliency Map Verbalization: Comparing Feature Importance Representations from Model-free and Instruction-based Methods ACL 2023
Saliency maps can explain a neural model's predictions by identifying important input features. They are difficult to interpret for laypeople, especially for instances with many features. In order to make them more accessible, we formalize the underexplored task of translating saliency maps into natural language and compare methods that address two key challenges of this approach -- what and how to verbalize. In both automatic and human evaluation setups, using token-level attributions from text classification tasks, we compare two novel methods (search-based and instruction-based verbalizations) against conventional feature importance representations (heatmap visualizations and extractive rationales), measuring simulatability, faithfulness, helpfulness and ease of understanding. Instructing GPT-3.5 to generate saliency map verbalizations yields plausible explanations which include associations, abstractive summarization and commonsense reasoning, achieving by far the highest human ratings, but they are not faithfully capturing numeric information and are inconsistent in their interpretation of the task. In comparison, our search-based, model-free verbalization approach efficiently completes templated verbalizations, is faithful by design, but falls short in helpfulness and simulatability. Our results suggest that saliency map verbalization makes feature attribution explanations more comprehensible and less cognitively challenging to humans than conventional representations.
comment: ACL 2023 Workshop on Natural Language Reasoning and Structured Explanations (NLRSE)
♻ ☆ Assessing Linguistic Generalisation in Language Models: A Dataset for Brazilian Portuguese
Much recent effort has been devoted to creating large-scale language models. Nowadays, the most prominent approaches are based on deep neural networks, such as BERT. However, they lack transparency and interpretability, and are often seen as black boxes. This affects not only their applicability in downstream tasks but also the comparability of different architectures or even of the same model trained using different corpora or hyperparameters. In this paper, we propose a set of intrinsic evaluation tasks that inspect the linguistic information encoded in models developed for Brazilian Portuguese. These tasks are designed to evaluate how different language models generalise information related to grammatical structures and multiword expressions (MWEs), thus allowing for an assessment of whether the model has learned different linguistic phenomena. The dataset that was developed for these tasks is composed of a series of sentences with a single masked word and a cue phrase that helps in narrowing down the context. This dataset is divided into MWEs and grammatical structures, and the latter is subdivided into 6 tasks: impersonal verbs, subject agreement, verb agreement, nominal agreement, passive and connectors. The subset for MWEs was used to test BERTimbau Large, BERTimbau Base and mBERT. For the grammatical structures, we used only BERTimbau Large, because it yielded the best results in the MWE task.
comment: This is the original manuscript that was submitted to LREV. The final version was published recently and can be found at: https://rdcu.be/ddEa6. Language Resources and Evaluation, https://doi.org/10.1007/s10579-023-09664-1
♻ ☆ Tractable Control for Autoregressive Language Generation
Despite the success of autoregressive large language models in text generation, it remains a major challenge to generate text that satisfies complex constraints: sampling from the conditional distribution ${\Pr}(\text{text} | \alpha)$ is intractable for even the simplest lexical constraints $\alpha$. To overcome this challenge, we propose to use tractable probabilistic models (TPMs) to impose lexical constraints in autoregressive text generation models, which we refer to as GeLaTo (Generating Language with Tractable Constraints). To demonstrate the effectiveness of this framework, we use distilled hidden Markov models, where we can efficiently compute ${\Pr}(\text{text} | \alpha)$, to guide autoregressive generation from GPT2. GeLaTo achieves state-of-the-art performance on challenging benchmarks for constrained text generation (e.g., CommonGen), beating various strong baselines by a large margin. Our work not only opens up new avenues for controlling large language models but also motivates the development of more expressive TPMs.
♻ ☆ SQL-PaLM: Improved Large Language Model Adaptation for Text-to-SQL
One impressive emergent capability of large language models (LLMs) is generation of code, including Structured Query Language (SQL) for databases. For the task of converting natural language text to SQL queries, Text-to-SQL, adaptation of LLMs is of paramount importance, both in in-context learning and fine-tuning settings, depending on the amount of adaptation data used. In this paper, we propose an LLM-based Text-to-SQL model SQL-PaLM, leveraging on PaLM-2, that pushes the state-of-the-art in both settings. Few-shot SQL-PaLM is based on an execution-based self-consistency prompting approach designed for Text-to-SQL, and achieves 77.3% in test-suite accuracy on Spider, which to our best knowledge is the first to outperform previous state-of-the-art with fine-tuning by a significant margin, 4%. Furthermore, we demonstrate that the fine-tuned SQL-PALM outperforms it further by another 1%. Towards applying SQL-PaLM to real-world scenarios we further evaluate its robustness on other challenging variants of Spider and demonstrate the superior generalization capability of SQL-PaLM. In addition, via extensive case studies, we demonstrate the impressive intelligent capabilities and various success enablers of LLM-based Text-to-SQL.
comment: 16 pages
♻ ☆ Test-Time Training on Nearest Neighbors for Large Language Models
Many recent efforts aim to augment language models with relevant information retrieved from a database at test time. We avoid the need for prompt engineering by directly fine-tuning the model on data retrieved at test time using its standard training setup. For this purpose, we build a large-scale distributed nearest neighbor index based on text embeddings of the Pile dataset. Given a query to a language model, our system retrieves the neighbors of the query and fine-tunes the model on the text data corresponding to those neighbors. Surprisingly, retrieving and training on as few as 20 neighbors, each for only one gradient iteration, drastically improves performance across more than twenty language modeling tasks in the Pile benchmark. For example, test-time training significantly narrows the performance gap between a small GPT2 model and a GPTNeo model, more than ten times larger, that was specifically trained to convergence on the Pile. Sufficient index quality and size, however, are important. Our work establishes a valuable first baseline for implementing test-time training in the context of large language models, opening the door to numerous promising research avenues.
comment: Corrected Figure 8. Code repository here: https://github.com/socialfoundations/tttlm
♻ ☆ Neural Natural Language Processing for Long Texts: A Survey of the State-of-the-Art
The adoption of Deep Neural Networks (DNNs) has greatly benefited Natural Language Processing (NLP) during the past decade. However, the demands of long document analysis are quite different from those of shorter texts, while the ever increasing size of documents uploaded on-line renders automated understanding of long texts a critical area of research. This article has two goals: a) it overviews the relevant neural building blocks, thus serving as a short tutorial, and b) it surveys the state-of-the-art in long document NLP, mainly focusing on two central tasks: document classification and document summarization. Sentiment analysis for long texts is also covered, since it is typically treated as a particular case of document classification. Thus, this article concerns document-level analysis. It discusses the main challenges and issues of long document NLP, along with the current solutions. Finally, the relevant, publicly available, annotated datasets are presented, in order to facilitate further research.
comment: 51 pages, 2 figures, 168 citations
♻ ☆ CADGE: Context-Aware Dialogue Generation Enhanced with Graph-Structured Knowledge Aggregation
Commonsense knowledge is crucial to many natural language processing tasks. Existing works usually incorporate graph knowledge with conventional graph neural networks (GNNs), leading to the text and graph knowledge encoding processes being separated in a serial pipeline. We argue that these separate representation learning stages may be suboptimal for neural networks to learn the overall context contained in both types of input knowledge. In this paper, we propose a novel context-aware graph-attention model (Context-aware GAT), which can effectively incorporate global features of relevant knowledge graphs based on a context-enhanced knowledge aggregation process. Specifically, our framework leverages a novel representation learning approach to process heterogeneous features - combining flattened graph knowledge with text. To the best of our knowledge, this is the first attempt at hierarchically applying graph knowledge aggregation on a connected subgraph in addition to contextual information to support commonsense dialogue generation. This framework shows superior performance compared to conventional GNN-based language frameworks. Both automatic and human evaluation demonstrates that our proposed model has significant performance uplifts over state-of-the-art baselines.
comment: Submitted to KBS
♻ ☆ TwistList: Resources and Baselines for Tongue Twister Generation
Previous work in phonetically-grounded language generation has mainly focused on domains such as lyrics and poetry. In this paper, we present work on the generation of tongue twisters - a form of language that is required to be phonetically conditioned to maximise sound overlap, whilst maintaining semantic consistency with an input topic, and still being grammatically correct. We present \textbf{TwistList}, a large annotated dataset of tongue twisters, consisting of 2.1K+ human-authored examples. We additionally present several benchmark systems (referred to as TwisterMisters) for the proposed task of tongue twister generation, including models that both do and do not require training on in-domain data. We present the results of automatic and human evaluation to demonstrate the performance of existing mainstream pre-trained models in this task with limited (or no) task specific training and data, and no explicit phonetic knowledge. We find that the task of tongue twister generation is challenging for models under these conditions, yet some models are still capable of generating acceptable examples of this language type.
♻ ☆ Early Discovery of Emerging Entities in Persian Twitter with Semantic Similarity
Discovering emerging entities (EEs) is the problem of finding entities before their establishment. These entities can be critical for individuals, companies, and governments. Many of these entities can be discovered on social media platforms, e.g. Twitter. These identities have been the spot of research in academia and industry in recent years. Similar to any machine learning problem, data availability is one of the major challenges in this problem. This paper proposes EEPT. That is an online clustering method able to discover EEs without any need for training on a dataset. Additionally, due to the lack of a proper evaluation metric, this paper uses a new metric to evaluate the results. The results show that EEPT is promising and finds significant entities before their establishment.
♻ ☆ Global Contrastive Batch Sampling via Optimization on Sample Permutations ICML 2023
Contrastive Learning has recently achieved state-of-the-art performance in a wide range of tasks. Many contrastive learning approaches use mined hard negatives to make batches more informative during training but these approaches are inefficient as they increase epoch length proportional to the number of mined negatives and require frequent updates of nearest neighbor indices or mining from recent batches. In this work, we provide an alternative to hard negative mining, Global Contrastive Batch Sampling (GCBS), an efficient approximation to the batch assignment problem that upper bounds the gap between the global and training losses, $\mathcal{L}^{Global} - \mathcal{L}^{Train}$, in contrastive learning settings. Through experimentation we find GCBS improves state-of-the-art performance in sentence embedding and code-search tasks. Additionally, GCBS is easy to implement as it requires only a few additional lines of code, does not maintain external data structures such as nearest neighbor indices, is more computationally efficient than the most minimal hard negative mining approaches, and makes no changes to the model being trained.
comment: ICML 2023; 21 pages, 7 figures
♻ ☆ bgGLUE: A Bulgarian General Language Understanding Evaluation Benchmark ACL 2023
We present bgGLUE(Bulgarian General Language Understanding Evaluation), a benchmark for evaluating language models on Natural Language Understanding (NLU) tasks in Bulgarian. Our benchmark includes NLU tasks targeting a variety of NLP problems (e.g., natural language inference, fact-checking, named entity recognition, sentiment analysis, question answering, etc.) and machine learning tasks (sequence labeling, document-level classification, and regression). We run the first systematic evaluation of pre-trained language models for Bulgarian, comparing and contrasting results across the nine tasks in the benchmark. The evaluation results show strong performance on sequence labeling tasks, but there is a lot of room for improvement for tasks that require more complex reasoning. We make bgGLUE publicly available together with the fine-tuning and the evaluation code, as well as a public leaderboard at https://bgglue.github.io/, and we hope that it will enable further advancements in developing NLU models for Bulgarian.
comment: Accepted to ACL 2023 (Main Conference)
♻ ☆ Learning Multi-Step Reasoning by Solving Arithmetic Tasks ACL 2023
Mathematical reasoning is regarded as a necessary ability for Language Models (LMs). Recent works demonstrate large LMs' impressive performance in solving math problems. The success is attributed to their Chain-of-Thought (CoT) reasoning abilities, i.e., the ability to decompose complex questions into step-by-step reasoning chains, but such ability seems only to emerge from models with abundant parameters. This work investigates how to incorporate relatively small LMs with the capabilities of multi-step reasoning. We propose to inject such abilities by continually pre-training LMs on a synthetic dataset MsAT which is composed of Multi-step Arithmetic Tasks. Our experiments on four math word problem datasets show the effectiveness of the proposed method in enhancing LMs' math reasoning abilities.
comment: ACL 2023. Code and data are available at https://github.com/TianduoWang/MsAT
♻ ☆ DataFinder: Scientific Dataset Recommendation from Natural Language Descriptions ACL 2023
Modern machine learning relies on datasets to develop and validate research ideas. Given the growth of publicly available data, finding the right dataset to use is increasingly difficult. Any research question imposes explicit and implicit constraints on how well a given dataset will enable researchers to answer this question, such as dataset size, modality, and domain. We operationalize the task of recommending datasets given a short natural language description of a research idea, to help people find relevant datasets for their needs. Dataset recommendation poses unique challenges as an information retrieval problem; datasets are hard to directly index for search and there are no corpora readily available for this task. To facilitate this task, we build the DataFinder Dataset which consists of a larger automatically-constructed training set (17.5K queries) and a smaller expert-annotated evaluation set (392 queries). Using this data, we compare various information retrieval algorithms on our test set and present a superior bi-encoder retriever for text-based dataset recommendation. This system, trained on the DataFinder Dataset, finds more relevant search results than existing third-party dataset search engines. To encourage progress on dataset recommendation, we release our dataset and models to the public.
comment: To appear at ACL 2023. Code published at https://github.com/viswavi/datafinder
♻ ☆ MACSum: Controllable Summarization with Mixed Attributes ACL 2023
Controllable summarization allows users to generate customized summaries with specified attributes. However, due to the lack of designated annotations of controlled summaries, existing works have to craft pseudo datasets by adapting generic summarization benchmarks. Furthermore, most research focuses on controlling single attributes individually (e.g., a short summary or a highly abstractive summary) rather than controlling a mix of attributes together (e.g., a short and highly abstractive summary). In this paper, we propose MACSum, the first human-annotated summarization dataset for controlling mixed attributes. It contains source texts from two domains, news articles and dialogues, with human-annotated summaries controlled by five designed attributes (Length, Extractiveness, Specificity, Topic, and Speaker). We propose two simple and effective parameter-efficient approaches for the new task of mixed controllable summarization based on hard prompt tuning and soft prefix tuning. Results and analysis demonstrate that hard prompt models yield the best performance on all metrics and human evaluations. However, mixed-attribute control is still challenging for summarization tasks. Our dataset and code are available at https://github.com/psunlpgroup/MACSum.
comment: TACL 2023
♻ ☆ ChatGPT Informed Graph Neural Network for Stock Movement Prediction
ChatGPT has demonstrated remarkable capabilities across various natural language processing (NLP) tasks. However, its potential for inferring dynamic network structures from temporal textual data, specifically financial news, remains an unexplored frontier. In this research, we introduce a novel framework that leverages ChatGPT's graph inference capabilities to enhance Graph Neural Networks (GNN). Our framework adeptly extracts evolving network structures from textual data, and incorporates these networks into graph neural networks for subsequent predictive tasks. The experimental results from stock movement forecasting indicate our model has consistently outperformed the state-of-the-art Deep Learning-based benchmarks. Furthermore, the portfolios constructed based on our model's outputs demonstrate higher annualized cumulative returns, alongside reduced volatility and maximum drawdown. This superior performance highlights the potential of ChatGPT for text-based network inferences and underscores its promising implications for the financial sector.
comment: Under Review. 10 pages, 2 figures
♻ ☆ Deductive Verification of Chain-of-Thought Reasoning
Large Language Models (LLMs) significantly benefit from Chain-of-Thought (CoT) prompting in performing various reasoning tasks. While CoT allows models to produce more comprehensive reasoning processes, its emphasis on intermediate reasoning steps can inadvertently introduce hallucinations and accumulated errors, thereby limiting models' ability to solve complex reasoning tasks. Inspired by how humans engage in careful and meticulous deductive logical reasoning processes to solve tasks, we seek to enable language models to perform explicit and rigorous deductive reasoning, and also ensure the trustworthiness of their reasoning process through self-verification. However, directly verifying the validity of an entire deductive reasoning process is challenging, even with advanced models like ChatGPT. In light of this, we propose to decompose a reasoning verification process into a series of step-by-step subprocesses, each only receiving their necessary context and premises. To facilitate this procedure, we propose Natural Program, a natural language-based deductive reasoning format. Our approach enables models to generate precise reasoning steps where subsequent steps are more rigorously grounded on prior steps. It also empowers language models to carry out reasoning self-verification in a step-by-step manner. By integrating this verification process into each deductive reasoning stage, we significantly enhance the rigor and trustfulness of generated reasoning steps. Along this process, we also improve the answer correctness on complex reasoning tasks. Code will be released at https://github.com/lz1oceani/verify_cot.
♻ ☆ Inference-Time Intervention: Eliciting Truthful Answers from a Language Model
We introduce Inference-Time Intervention (ITI), a technique designed to enhance the truthfulness of large language models (LLMs). ITI operates by shifting model activations during inference, following a set of directions across a limited number of attention heads. This intervention significantly improves the performance of LLaMA models on the TruthfulQA benchmark. On an instruction-finetuned LLaMA called Alpaca, ITI improves its truthfulness from 32.5% to 65.1%. We identify a tradeoff between truthfulness and helpfulness and demonstrate how to balance it by tuning the intervention strength. ITI is minimally invasive and computationally inexpensive. Moreover, the technique is data efficient: while approaches like RLHF require extensive annotations, ITI locates truthful directions using only few hundred examples. Our findings suggest that LLMs may have an internal representation of the likelihood of something being true, even as they produce falsehoods on the surface.
comment: code: https://github.com/likenneth/honest_llama
Computer Vision and Pattern Recognition 98
☆ GP-UNIT: Generative Prior for Versatile Unsupervised Image-to-Image Translation
Recent advances in deep learning have witnessed many successful unsupervised image-to-image translation models that learn correspondences between two visual domains without paired data. However, it is still a great challenge to build robust mappings between various domains especially for those with drastic visual discrepancies. In this paper, we introduce a novel versatile framework, Generative Prior-guided UNsupervised Image-to-image Translation (GP-UNIT), that improves the quality, applicability and controllability of the existing translation models. The key idea of GP-UNIT is to distill the generative prior from pre-trained class-conditional GANs to build coarse-level cross-domain correspondences, and to apply the learned prior to adversarial translations to excavate fine-level correspondences. With the learned multi-level content correspondences, GP-UNIT is able to perform valid translations between both close domains and distant domains. For close domains, GP-UNIT can be conditioned on a parameter to determine the intensity of the content correspondences during translation, allowing users to balance between content and style consistency. For distant domains, semi-supervised learning is explored to guide GP-UNIT to discover accurate semantic correspondences that are hard to learn solely from the appearance. We validate the superiority of GP-UNIT over state-of-the-art translation models in robust, high-quality and diversified translations between various domains through extensive experiments.
comment: Accepted by IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI). Code: https://github.com/williamyang1991/GP-UNIT Project page: https://www.mmlab-ntu.com/project/gpunit/. arXiv admin note: substantial text overlap with arXiv:2204.03641
☆ Contrastive Lift: 3D Object Instance Segmentation by Slow-Fast Contrastive Fusion
Instance segmentation in 3D is a challenging task due to the lack of large-scale annotated datasets. In this paper, we show that this task can be addressed effectively by leveraging instead 2D pre-trained models for instance segmentation. We propose a novel approach to lift 2D segments to 3D and fuse them by means of a neural field representation, which encourages multi-view consistency across frames. The core of our approach is a slow-fast clustering objective function, which is scalable and well-suited for scenes with a large number of objects. Unlike previous approaches, our method does not require an upper bound on the number of objects or object tracking across frames. To demonstrate the scalability of the slow-fast clustering, we create a new semi-realistic dataset called the Messy Rooms dataset, which features scenes with up to 500 objects per scene. Our approach outperforms the state-of-the-art on challenging scenes from the ScanNet, Hypersim, and Replica datasets, as well as on our newly created Messy Rooms dataset, demonstrating the effectiveness and scalability of our slow-fast clustering method.
☆ Designing a Better Asymmetric VQGAN for StableDiffusion
StableDiffusion is a revolutionary text-to-image generator that is causing a stir in the world of image generation and editing. Unlike traditional methods that learn a diffusion model in pixel space, StableDiffusion learns a diffusion model in the latent space via a VQGAN, ensuring both efficiency and quality. It not only supports image generation tasks, but also enables image editing for real images, such as image inpainting and local editing. However, we have observed that the vanilla VQGAN used in StableDiffusion leads to significant information loss, causing distortion artifacts even in non-edited image regions. To this end, we propose a new asymmetric VQGAN with two simple designs. Firstly, in addition to the input from the encoder, the decoder contains a conditional branch that incorporates information from task-specific priors, such as the unmasked image region in inpainting. Secondly, the decoder is much heavier than the encoder, allowing for more detailed recovery while only slightly increasing the total inference cost. The training cost of our asymmetric VQGAN is cheap, and we only need to retrain a new asymmetric decoder while keeping the vanilla VQGAN encoder and StableDiffusion unchanged. Our asymmetric VQGAN can be widely used in StableDiffusion-based inpainting and local editing methods. Extensive experiments demonstrate that it can significantly improve the inpainting and editing performance, while maintaining the original text-to-image capability. The code is available at \url{https://github.com/buxiangzhiren/Asymmetric_VQGAN}.
comment: code is available at https://github.com/buxiangzhiren/Asymmetric_VQGAN
☆ Yet Another Algorithm for Supervised Principal Component Analysis: Supervised Linear Centroid-Encoder
We propose a new supervised dimensionality reduction technique called Supervised Linear Centroid-Encoder (SLCE), a linear counterpart of the nonlinear Centroid-Encoder (CE) \citep{ghosh2022supervised}. SLCE works by mapping the samples of a class to its class centroid using a linear transformation. The transformation is a projection that reconstructs a point such that its distance from the corresponding class centroid, i.e., centroid-reconstruction loss, is minimized in the ambient space. We derive a closed-form solution using an eigendecomposition of a symmetric matrix. We did a detailed analysis and presented some crucial mathematical properties of the proposed approach. %We also provide an iterative solution approach based solving the optimization problem using a descent method. We establish a connection between the eigenvalues and the centroid-reconstruction loss. In contrast to Principal Component Analysis (PCA) which reconstructs a sample in the ambient space, the transformation of SLCE uses the instances of a class to rebuild the corresponding class centroid. Therefore the proposed method can be considered a form of supervised PCA. Experimental results show the performance advantage of SLCE over other supervised methods.
comment: A novel algorithm for supervised PCA. 22 pages (including 2 reference pages), 8 figures and mathematical analysis of the proposed algorithm. The article is under review now
☆ Align, Distill, and Augment Everything All at Once for Imbalanced Semi-Supervised Learning
Addressing the class imbalance in long-tailed semi-supervised learning (SSL) poses a few significant challenges stemming from differences between the marginal distributions of unlabeled data and the labeled data, as the former is often unknown and potentially distinct from the latter. The first challenge is to avoid biasing the pseudo-labels towards an incorrect distribution, such as that of the labeled data or a balanced distribution, during training. However, we still wish to ensure a balanced unlabeled distribution during inference, which is the second challenge. To address both of these challenges, we propose a three-faceted solution: a flexible distribution alignment that progressively aligns the classifier from a dynamically estimated unlabeled prior towards a balanced distribution, a soft consistency regularization that exploits underconfident pseudo-labels discarded by threshold-based methods, and a schema for expanding the unlabeled set with input data from the labeled partition. This last facet comes in as a response to the commonly-overlooked fact that disjoint partitions of labeled and unlabeled data prevent the benefits of strong data augmentation on the labeled set. Our overall framework requires no additional training cycles, so it will align, distill, and augment everything all at once (ADALLO). Our extensive evaluations of ADALLO on imbalanced SSL benchmark datasets, including CIFAR10-LT, CIFAR100-LT, and STL10-LT with varying degrees of class imbalance, amount of labeled data, and distribution mismatch, demonstrate significant improvements in the performance of imbalanced SSL under large distribution mismatch, as well as competitiveness with state-of-the-art methods when the labeled and unlabeled data follow the same marginal distribution. Our code will be released upon paper acceptance.
comment: Under review, 12 pages
☆ ARTIC3D: Learning Robust Articulated 3D Shapes from Noisy Web Image Collections
Estimating 3D articulated shapes like animal bodies from monocular images is inherently challenging due to the ambiguities of camera viewpoint, pose, texture, lighting, etc. We propose ARTIC3D, a self-supervised framework to reconstruct per-instance 3D shapes from a sparse image collection in-the-wild. Specifically, ARTIC3D is built upon a skeleton-based surface representation and is further guided by 2D diffusion priors from Stable Diffusion. First, we enhance the input images with occlusions/truncation via 2D diffusion to obtain cleaner mask estimates and semantic features. Second, we perform diffusion-guided 3D optimization to estimate shape and texture that are of high-fidelity and faithful to input images. We also propose a novel technique to calculate more stable image-level gradients via diffusion models compared to existing alternatives. Finally, we produce realistic animations by fine-tuning the rendered shape and texture under rigid part transformations. Extensive evaluations on multiple existing datasets as well as newly introduced noisy web image collections with occlusions and truncation demonstrate that ARTIC3D outputs are more robust to noisy images, higher quality in terms of shape and texture details, and more realistic when animated. Project page: https://chhankyao.github.io/artic3d/
comment: Project page: https://chhankyao.github.io/artic3d/
☆ ICON$^2$: Reliably Benchmarking Predictive Inequity in Object Detection CVPR 2023
As computer vision systems are being increasingly deployed at scale in high-stakes applications like autonomous driving, concerns about social bias in these systems are rising. Analysis of fairness in real-world vision systems, such as object detection in driving scenes, has been limited to observing predictive inequity across attributes such as pedestrian skin tone, and lacks a consistent methodology to disentangle the role of confounding variables e.g. does my model perform worse for a certain skin tone, or are such scenes in my dataset more challenging due to occlusion and crowds? In this work, we introduce ICON$^2$, a framework for robustly answering this question. ICON$^2$ leverages prior knowledge on the deficiencies of object detection systems to identify performance discrepancies across sub-populations, compute correlations between these potential confounders and a given sensitive attribute, and control for the most likely confounders to obtain a more reliable estimate of model bias. Using our approach, we conduct an in-depth study on the performance of object detection with respect to income from the BDD100K driving dataset, revealing useful insights.
comment: Accepted to CVPR 2023 SSAD Workshop
☆ Integrating Geometric Control into Text-to-Image Diffusion Models for High-Quality Detection Data Generation via Text Prompt
Diffusion models have attracted significant attention due to their remarkable ability to create content and generate data for tasks such as image classification. However, the usage of diffusion models to generate high-quality object detection data remains an underexplored area, where not only the image-level perceptual quality but also geometric conditions such as bounding boxes and camera views are essential. Previous studies have utilized either copy-paste synthesis or layout-to-image (L2I) generation with specifically designed modules to encode semantic layouts. In this paper, we propose GeoDiffusion, a simple framework that can flexibly translate various geometric conditions into text prompts and empower the pre-trained text-to-image (T2I) diffusion models for high-quality detection data generation. Unlike previous L2I methods, our GeoDiffusion is able to encode not only bounding boxes but also extra geometric conditions such as camera views in self-driving scenes. Extensive experiments demonstrate GeoDiffusion outperforms previous L2I methods while maintaining 4x training time faster. To the best of our knowledge, this is the first work to adopt diffusion models for layout-to-image generation with geometric conditions and demonstrate that L2I-generated images can be beneficial for improving the performance of object detectors.
☆ MarineVRS: Marine Video Retrieval System with Explainability via Semantic Understanding
Building a video retrieval system that is robust and reliable, especially for the marine environment, is a challenging task due to several factors such as dealing with massive amounts of dense and repetitive data, occlusion, blurriness, low lighting conditions, and abstract queries. To address these challenges, we present MarineVRS, a novel and flexible video retrieval system designed explicitly for the marine domain. MarineVRS integrates state-of-the-art methods for visual and linguistic object representation to enable efficient and accurate search and analysis of vast volumes of underwater video data. In addition, unlike the conventional video retrieval system, which only permits users to index a collection of images or videos and search using a free-form natural language sentence, our retrieval system includes an additional Explainability module that outputs the segmentation masks of the objects that the input query referred to. This feature allows users to identify and isolate specific objects in the video footage, leading to more detailed analysis and understanding of their behavior and movements. Finally, with its adaptability, explainability, accuracy, and scalability, MarineVRS is a powerful tool for marine researchers and scientists to efficiently and accurately process vast amounts of data and gain deeper insights into the behavior and movements of marine species.
comment: Accepted to OCEANS 2023 Limerick. Website: https://marinevrs.hkustvgd.com/
☆ A Dataset for Deep Learning-based Bone Structure Analyses in Total Hip Arthroplasty
Total hip arthroplasty (THA) is a widely used surgical procedure in orthopedics. For THA, it is of clinical significance to analyze the bone structure from the CT images, especially to observe the structure of the acetabulum and femoral head, before the surgical procedure. For such bone structure analyses, deep learning technologies are promising but require high-quality labeled data for the learning, while the data labeling is costly. We address this issue and propose an efficient data annotation pipeline for producing a deep learning-oriented dataset. Our pipeline consists of non-learning-based bone extraction (BE) and acetabulum and femoral head segmentation (AFS) and active-learning-based annotation refinement (AAR). For BE we use the classic graph-cut algorithm. For AFS we propose an improved algorithm, including femoral head boundary localization using first-order and second-order gradient regularization, line-based non-maximum suppression, and anatomy prior-based femoral head extraction. For AAR, we refine the algorithm-produced pseudo labels with the help of trained deep models: we measure the uncertainty based on the disagreement between the original pseudo labels and the deep model predictions, and then find out the samples with the largest uncertainty to ask for manual labeling. Using the proposed pipeline, we construct a large-scale bone structure analyses dataset from more than 300 clinical and diverse CT scans. We perform careful manual labeling for the test set of our data. We then benchmark multiple state-of-the art deep learning-based methods of medical image segmentation using the training and test sets of our data. The extensive experimental results validate the efficacy of the proposed data annotation pipeline. The dataset, related codes and models will be publicly available at https://github.com/hitachinsk/THA.
comment: 16 pages, 17 figures
☆ PhenoBench -- A Large Dataset and Benchmarks for Semantic Image Interpretation in the Agricultural Domain
The production of food, feed, fiber, and fuel is a key task of agriculture. Especially crop production has to cope with a multitude of challenges in the upcoming decades caused by a growing world population, climate change, the need for sustainable production, lack of skilled workers, and generally the limited availability of arable land. Vision systems could help cope with these challenges by offering tools to make better and more sustainable field management decisions and support the breeding of new varieties of crops by allowing temporally dense and reproducible measurements. Recently, tackling perception tasks in the agricultural domain got increasing interest in the computer vision and robotics community since agricultural robotics are one promising solution for coping with the lack of workers and enable a more sustainable agricultural production at the same time. While large datasets and benchmarks in other domains are readily available and have enabled significant progress toward more reliable vision systems, agricultural datasets and benchmarks are comparably rare. In this paper, we present a large dataset and benchmarks for the semantic interpretation of images of real agricultural fields. Our dataset recorded with a UAV provides high-quality, dense annotations of crops and weeds, but also fine-grained labels of crop leaves at the same time, which enable the development of novel algorithms for visual perception in the agricultural domain. Together with the labeled data, we provide novel benchmarks for evaluating different visual perception tasks on a hidden test set comprised of different fields: known fields covered by the training data and a completely unseen field. The tasks cover semantic segmentation, panoptic segmentation of plants, leaf instance segmentation, detection of plants and leaves, and hierarchical panoptic segmentation for jointly identifying plants and leaves.
☆ On the Design Fundamentals of Diffusion Models: A Survey
Diffusion models are generative models, which gradually add and remove noise to learn the underlying distribution of training data for data generation. The components of diffusion models have gained significant attention with many design choices proposed. Existing reviews have primarily focused on higher-level solutions, thereby covering less on the design fundamentals of components. This study seeks to address this gap by providing a comprehensive and coherent review on component-wise design choices in diffusion models. Specifically, we organize this review according to their three key components, namely the forward process, the reverse process, and the sampling procedure. This allows us to provide a fine-grained perspective of diffusion models, benefiting future studies in the analysis of individual components, the applicability of design choices, and the implementation of diffusion models.
☆ NeMO: Neural Map Growing System for Spatiotemporal Fusion in Bird's-Eye-View and BDD-Map Benchmark
Vision-centric Bird's-Eye View (BEV) representation is essential for autonomous driving systems (ADS). Multi-frame temporal fusion which leverages historical information has been demonstrated to provide more comprehensive perception results. While most research focuses on ego-centric maps of fixed settings, long-range local map generation remains less explored. This work outlines a new paradigm, named NeMO, for generating local maps through the utilization of a readable and writable big map, a learning-based fusion module, and an interaction mechanism between the two. With an assumption that the feature distribution of all BEV grids follows an identical pattern, we adopt a shared-weight neural network for all grids to update the big map. This paradigm supports the fusion of longer time series and the generation of long-range BEV local maps. Furthermore, we release BDD-Map, a BDD100K-based dataset incorporating map element annotations, including lane lines, boundaries, and pedestrian crossing. Experiments on the NuScenes and BDD-Map datasets demonstrate that NeMO outperforms state-of-the-art map segmentation methods. We also provide a new scene-level BEV map evaluation setting along with the corresponding baseline for a more comprehensive comparison.
☆ Multimodal Learning Without Labeled Multimodal Data: Guarantees and Applications
In many machine learning systems that jointly learn from multiple modalities, a core research question is to understand the nature of multimodal interactions: the emergence of new task-relevant information during learning from both modalities that was not present in either alone. We study this challenge of interaction quantification in a semi-supervised setting with only labeled unimodal data and naturally co-occurring multimodal data (e.g., unlabeled images and captions, video and corresponding audio) but when labeling them is time-consuming. Using a precise information-theoretic definition of interactions, our key contributions are the derivations of lower and upper bounds to quantify the amount of multimodal interactions in this semi-supervised setting. We propose two lower bounds based on the amount of shared information between modalities and the disagreement between separately trained unimodal classifiers, and derive an upper bound through connections to approximate algorithms for min-entropy couplings. We validate these estimated bounds and show how they accurately track true interactions. Finally, two semi-supervised multimodal applications are explored based on these theoretical results: (1) analyzing the relationship between multimodal performance and estimated interactions, and (2) self-supervised learning that embraces disagreement between modalities beyond agreement as is typically done.
comment: Code available at: https://github.com/pliang279/PID
☆ ContriMix: Unsupervised disentanglement of content and attribute for domain generalization in microscopy image analysis
Domain generalization is critical for real-world applications of machine learning models to microscopy images, including histopathology and fluorescence imaging. Artifacts in histopathology arise through a complex combination of factors relating to tissue collection and laboratory processing, as well as factors intrinsic to patient samples. In fluorescence imaging, these artifacts stem from variations across experimental batches. The complexity and subtlety of these artifacts make the enumeration of data domains intractable. Therefore, augmentation-based methods of domain generalization that require domain identifiers and manual fine-tuning are inadequate in this setting. To overcome this challenge, we introduce ContriMix, a domain generalization technique that learns to generate synthetic images by disentangling and permuting the biological content ("content") and technical variations ("attributes") in microscopy images. ContriMix does not rely on domain identifiers or handcrafted augmentations and makes no assumptions about the input characteristics of images. We assess the performance of ContriMix on two pathology datasets (Camelyon17-WILDS and a prostate cell classification dataset) and one fluorescence microscopy dataset (RxRx1-WILDS). ContriMix outperforms current state-of-the-art methods in all datasets, motivating its usage for microscopy image analysis in real-world settings where domain information is hard to come by.
☆ Cross-attention learning enables real-time nonuniform rotational distortion correction in OCT
Nonuniform rotational distortion (NURD) correction is vital for endoscopic optical coherence tomography (OCT) imaging and its functional extensions, such as angiography and elastography. Current NURD correction methods require time-consuming feature tracking or cross-correlation calculations and thus sacrifice temporal resolution. Here we propose a cross-attention learning method for the NURD correction in OCT. Our method is inspired by the recent success of the self-attention mechanism in natural language processing and computer vision. By leveraging its ability to model long-range dependencies, we can directly obtain the correlation between OCT A-lines at any distance, thus accelerating the NURD correction. We develop an end-to-end stacked cross-attention network and design three types of optimization constraints. We compare our method with two traditional feature-based methods and a CNN-based method, on two publicly-available endoscopic OCT datasets and a private dataset collected on our home-built endoscopic OCT system. Our method achieved a $\sim3\times$ speedup to real time ($26\pm 3$ fps), and superior correction performance.
comment: 4 pages, 6 figures
☆ Improving neural network representations using human similarity judgments
Deep neural networks have reached human-level performance on many computer vision tasks. However, the objectives used to train these networks enforce only that similar images are embedded at similar locations in the representation space, and do not directly constrain the global structure of the resulting space. Here, we explore the impact of supervising this global structure by linearly aligning it with human similarity judgments. We find that a naive approach leads to large changes in local representational structure that harm downstream performance. Thus, we propose a novel method that aligns the global structure of representations while preserving their local structure. This global-local transform considerably improves accuracy across a variety of few-shot learning and anomaly detection tasks. Our results indicate that human visual representations are globally organized in a way that facilitates learning from few examples, and incorporating this global structure into neural network representations improves performance on downstream tasks.
☆ Defocus to focus: Photo-realistic bokeh rendering by fusing defocus and radiance priors
We consider the problem of realistic bokeh rendering from a single all-in-focus image. Bokeh rendering mimics aesthetic shallow depth-of-field (DoF) in professional photography, but these visual effects generated by existing methods suffer from simple flat background blur and blurred in-focus regions, giving rise to unrealistic rendered results. In this work, we argue that realistic bokeh rendering should (i) model depth relations and distinguish in-focus regions, (ii) sustain sharp in-focus regions, and (iii) render physically accurate Circle of Confusion (CoC). To this end, we present a Defocus to Focus (D2F) framework to learn realistic bokeh rendering by fusing defocus priors with the all-in-focus image and by implementing radiance priors in layered fusion. Since no depth map is provided, we introduce defocus hallucination to integrate depth by learning to focus. The predicted defocus map implies the blur amount of bokeh and is used to guide weighted layered rendering. In layered rendering, we fuse images blurred by different kernels based on the defocus map. To increase the reality of the bokeh, we adopt radiance virtualization to simulate scene radiance. The scene radiance used in weighted layered rendering reassigns weights in the soft disk kernel to produce the CoC. To ensure the sharpness of in-focus regions, we propose to fuse upsampled bokeh images and original images. We predict the initial fusion mask from our defocus map and refine the mask with a deep network. We evaluate our model on a large-scale bokeh dataset. Extensive experiments show that our approach is capable of rendering visually pleasing bokeh effects in complex scenes. In particular, our solution receives the runner-up award in the AIM 2020 Rendering Realistic Bokeh Challenge.
comment: Published at Information Fusion 2023 https://www.sciencedirect.com/science/article/pii/S1566253522001221
☆ Rewarded soups: towards Pareto-optimal alignment by interpolating weights fine-tuned on diverse rewards
Foundation models are first pre-trained on vast unsupervised datasets and then fine-tuned on labeled data. Reinforcement learning, notably from human feedback (RLHF), can further align the network with the intended usage. Yet the imperfections in the proxy reward may hinder the training and lead to suboptimal results; the diversity of objectives in real-world tasks and human opinions exacerbate the issue. This paper proposes embracing the heterogeneity of diverse rewards by following a multi-policy strategy. Rather than focusing on a single a priori reward, we aim for Pareto-optimal generalization across the entire space of preferences. To this end, we propose rewarded soup, first specializing multiple networks independently (one for each proxy reward) and then interpolating their weights linearly. This succeeds empirically because we show that the weights remain linearly connected when fine-tuned on diverse rewards from a shared pre-trained initialization. We demonstrate the effectiveness of our approach for text-to-text (summarization, Q&A, helpful assistant, review), text-image (image captioning, text-to-image generation, visual grounding, VQA), and control (locomotion) tasks. We hope to enhance the alignment of deep models, and how they interact with the world in all its diversity.
☆ FoSp: Focus and Separation Network for Early Smoke Segmentation
Early smoke segmentation (ESS) enables the accurate identification of smoke sources, facilitating the prompt extinguishing of fires and preventing large-scale gas leaks. But ESS poses greater challenges than conventional object and regular smoke segmentation due to its small scale and transparent appearance, which can result in high miss detection rate and low precision. To address these issues, a Focus and Separation Network (FoSp) is proposed. We first introduce a Focus module employing bidirectional cascade which guides low-resolution and high-resolution features towards mid-resolution to locate and determine the scope of smoke, reducing the miss detection rate. Next, we propose a Separation module that separates smoke images into a pure smoke foreground and a smoke-free background, enhancing the contrast between smoke and background fundamentally, improving segmentation precision. Finally, a Domain Fusion module is developed to integrate the distinctive features of the two modules which can balance recall and precision to achieve high F_beta. Futhermore, to promote the development of ESS, we introduce a high-quality real-world dataset called SmokeSeg, which contains more small and transparent smoke than the existing datasets. Experimental results show that our model achieves the best performance on three available datasets: SYN70K (mIoU: 83.00%), SMOKE5K (F_beta: 81.6%) and SmokeSeg (F_beta: 72.05%). Especially, our FoSp outperforms SegFormer by 7.71% (F_beta) for early smoke segmentation on SmokeSeg.
☆ Improving Diffusion-based Image Translation using Asymmetric Gradient Guidance
Diffusion models have shown significant progress in image translation tasks recently. However, due to their stochastic nature, there's often a trade-off between style transformation and content preservation. Current strategies aim to disentangle style and content, preserving the source image's structure while successfully transitioning from a source to a target domain under text or one-shot image conditions. Yet, these methods often require computationally intense fine-tuning of diffusion models or additional neural networks. To address these challenges, here we present an approach that guides the reverse process of diffusion sampling by applying asymmetric gradient guidance. This results in quicker and more stable image manipulation for both text-guided and image-guided image translation. Our model's adaptability allows it to be implemented with both image- and latent-diffusion models. Experiments show that our method outperforms various state-of-the-art models in image translation tasks.
☆ M$^3$IT: A Large-Scale Dataset towards Multi-Modal Multilingual Instruction Tuning
Instruction tuning has significantly advanced large language models (LLMs) such as ChatGPT, enabling them to align with human instructions across diverse tasks. However, progress in open vision-language models (VLMs) has been limited due to the scarcity of high-quality instruction datasets. To tackle this challenge and promote research in the vision-language field, we introduce the Multi-Modal, Multilingual Instruction Tuning (M$^3$IT) dataset, designed to optimize VLM alignment with human instructions. Our M$^3$IT dataset comprises 40 carefully curated datasets, including 2.4 million instances and 400 manually written task instructions, reformatted into a vision-to-text structure. Key tasks are translated into 80 languages with an advanced translation system, ensuring broader accessibility. M$^3$IT surpasses previous datasets regarding task coverage, instruction number and instance scale. Moreover, we develop Ying-VLM, a VLM model trained on our M$^3$IT dataset, showcasing its potential to answer complex questions requiring world knowledge, generalize to unseen video tasks, and comprehend unseen instructions in Chinese. To encourage further research, we have open-sourced both the dataset and trained models.
comment: Dataset available at: https://huggingface.co/MMInstruction/M3IT
☆ SF-FSDA: Source-Free Few-Shot Domain Adaptive Object Detection with Efficient Labeled Data Factory
Domain adaptive object detection aims to leverage the knowledge learned from a labeled source domain to improve the performance on an unlabeled target domain. Prior works typically require the access to the source domain data for adaptation, and the availability of sufficient data on the target domain. However, these assumptions may not hold due to data privacy and rare data collection. In this paper, we propose and investigate a more practical and challenging domain adaptive object detection problem under both source-free and few-shot conditions, named as SF-FSDA. To overcome this problem, we develop an efficient labeled data factory based approach. Without accessing the source domain, the data factory renders i) infinite amount of synthesized target-domain like images, under the guidance of the few-shot image samples and text description from the target domain; ii) corresponding bounding box and category annotations, only demanding minimum human effort, i.e., a few manually labeled examples. On the one hand, the synthesized images mitigate the knowledge insufficiency brought by the few-shot condition. On the other hand, compared to the popular pseudo-label technique, the generated annotations from data factory not only get rid of the reliance on the source pretrained object detection model, but also alleviate the unavoidably pseudo-label noise due to domain shift and source-free condition. The generated dataset is further utilized to adapt the source pretrained object detection model, realizing the robust object detection under SF-FSDA. The experiments on different settings showcase that our proposed approach outperforms other state-of-the-art methods on SF-FSDA problem. Our codes and models will be made publicly available.
☆ Youku-mPLUG: A 10 Million Large-scale Chinese Video-Language Dataset for Pre-training and Benchmarks
To promote the development of Vision-Language Pre-training (VLP) and multimodal Large Language Model (LLM) in the Chinese community, we firstly release the largest public Chinese high-quality video-language dataset named Youku-mPLUG, which is collected from Youku, a well-known Chinese video-sharing website, with strict criteria of safety, diversity, and quality. Youku-mPLUG contains 10 million Chinese video-text pairs filtered from 400 million raw videos across a wide range of 45 diverse categories for large-scale pre-training. In addition, to facilitate a comprehensive evaluation of video-language models, we carefully build the largest human-annotated Chinese benchmarks covering three popular video-language tasks of cross-modal retrieval, video captioning, and video category classification. Youku-mPLUG can enable researchers to conduct more in-depth multimodal research and develop better applications in the future. Furthermore, we release popular video-language pre-training models, ALPRO and mPLUG-2, and our proposed modularized decoder-only model mPLUG-video pre-trained on Youku-mPLUG. Experiments show that models pre-trained on Youku-mPLUG gain up to 23.1% improvement in video category classification. Besides, mPLUG-video achieves a new state-of-the-art result on these benchmarks with 80.5% top-1 accuracy in video category classification and 68.9 CIDEr score in video captioning, respectively. Finally, we scale up mPLUG-video based on the frozen Bloomz with only 1.7% trainable parameters as Chinese multimodal LLM, and demonstrate impressive instruction and video understanding ability. The zero-shot instruction understanding experiment indicates that pretraining with Youku-mPLUG can enhance the ability to comprehend overall and detailed visual semantics, recognize scene text, and leverage open-domain knowledge.
comment: Working in progress
☆ Fine-Grained Visual Prompting
Vision-Language Models (VLMs), such as CLIP, have demonstrated impressive zero-shot transfer capabilities in image-level visual perception. However, these models have shown limited performance in instance-level tasks that demand precise localization and recognition. Previous works have suggested that incorporating visual prompts, such as colorful boxes or circles, can improve the ability of models to recognize objects of interest. Nonetheless, compared to language prompting, visual prompting designs are rarely explored. Existing approaches, which employ coarse visual cues such as colorful boxes or circles, often result in sub-optimal performance due to the inclusion of irrelevant and noisy pixels. In this paper, we carefully study the visual prompting designs by exploring more fine-grained markings, such as segmentation masks and their variations. In addition, we introduce a new zero-shot framework that leverages pixel-level annotations acquired from a generalist segmentation model for fine-grained visual prompting. Consequently, our investigation reveals that a straightforward application of blur outside the target mask, referred to as the Blur Reverse Mask, exhibits exceptional effectiveness. This proposed prompting strategy leverages the precise mask annotations to reduce focus on weakly related regions while retaining spatial coherence between the target and the surrounding background. Our Fine-Grained Visual Prompting (FGVP) demonstrates superior performance in zero-shot comprehension of referring expressions on the RefCOCO, RefCOCO+, and RefCOCOg benchmarks. It outperforms prior methods by an average margin of 3.0% to 4.6%, with a maximum improvement of 12.5% on the RefCOCO+ testA subset. The part detection experiments conducted on the PACO dataset further validate the preponderance of FGVP over existing visual prompting techniques. Code and models will be made available.
☆ An Overview of Challenges in Egocentric Text-Video Retrieval CVPR 2023
Text-video retrieval contains various challenges, including biases coming from diverse sources. We highlight some of them supported by illustrations to open a discussion. Besides, we address one of the biases, frame length bias, with a simple method which brings a very incremental but promising increase. We conclude with future directions.
comment: 4 pages, CVPR 2023 Joint Ego4D&EPIC Workshop, Extended Abstract
☆ ViDA: Homeostatic Visual Domain Adapter for Continual Test Time Adaptation
Since real-world machine systems are running in non-stationary and continually changing environments, Continual Test-Time Adaptation (CTTA) task is proposed to adapt the pre-trained model to continually changing target domains. Recently, existing methods mainly focus on model-based adaptation, which aims to leverage a self-training manner to extract the target domain knowledge. However, pseudo labels can be noisy and the updated model parameters are uncertain under dynamic data distributions, leading to error accumulation and catastrophic forgetting in the continual adaptation process. To tackle these challenges and maintain the model plasticity, we tactfully design a Visual Domain Adapter (ViDA) for CTTA, explicitly handling both domain-specific and domain-agnostic knowledge. Specifically, we first comprehensively explore the different domain representations of the adapters with trainable high and low-rank embedding space. Then we inject ViDAs into the pre-trained model, which leverages high-rank and low-rank prototypes to adapt the current domain distribution and maintain the continual domain-shared knowledge, respectively. To adapt to the various distribution shifts of each sample in target domains, we further propose a Homeostatic Knowledge Allotment (HKA) strategy, which adaptively merges knowledge from each ViDA with different rank prototypes. Extensive experiments conducted on four widely-used benchmarks demonstrate that our proposed method achieves state-of-the-art performance in both classification and segmentation CTTA tasks. In addition, our method can be regarded as a novel transfer paradigm and showcases promising results in zero-shot adaptation of foundation models to continual downstream tasks and distributions.
☆ Unpaired Deep Learning for Pharmacokinetic Parameter Estimation from Dynamic Contrast-Enhanced MRI
DCE-MRI provides information about vascular permeability and tissue perfusion through the acquisition of pharmacokinetic parameters. However, traditional methods for estimating these pharmacokinetic parameters involve fitting tracer kinetic models, which often suffer from computational complexity and low accuracy due to noisy arterial input function (AIF) measurements. Although some deep learning approaches have been proposed to tackle these challenges, most existing methods rely on supervised learning that requires paired input DCE-MRI and labeled pharmacokinetic parameter maps. This dependency on labeled data introduces significant time and resource constraints, as well as potential noise in the labels, making supervised learning methods often impractical. To address these limitations, here we present a novel unpaired deep learning method for estimating both pharmacokinetic parameters and the AIF using a physics-driven CycleGAN approach. Our proposed CycleGAN framework is designed based on the underlying physics model, resulting in a simpler architecture with a single generator and discriminator pair. Crucially, our experimental results indicate that our method, which does not necessitate separate AIF measurements, produces more reliable pharmacokinetic parameters than other techniques.
☆ CorrMatch: Label Propagation via Correlation Matching for Semi-Supervised Semantic Segmentation
In this paper, we present a simple but performant semi-supervised semantic segmentation approach, termed CorrMatch. Our goal is to mine more high-quality regions from the unlabeled images to leverage the unlabeled data more efficiently via consistency regularization. The key contributions of our CorrMatch are two novel and complementary strategies. First, we introduce an adaptive threshold updating strategy with a relaxed initialization to expand the high-quality regions. Furthermore, we propose to propagate high-confidence predictions through measuring the pairwise similarities between pixels. Despite its simplicity, we show that CorrMatch achieves great performance on popular semi-supervised semantic segmentation benchmarks. Taking the DeepLabV3+ framework with ResNet-101 backbone as our segmentation model, we receive a 76%+ mIoU score on the Pascal VOC 2012 segmentation benchmark with only 92 annotated images provided. We also achieve a consistent improvement over previous semi-supervised semantic segmentation models. Code will be made publicly available.
☆ Revising deep learning methods in parking lot occupancy detection
Parking guidance systems have recently become a popular trend as a part of the smart cities' paradigm of development. The crucial part of such systems is the algorithm allowing drivers to search for available parking lots across regions of interest. The classic approach to this task is based on the application of neural network classifiers to camera records. However, existing systems demonstrate a lack of generalization ability and appropriate testing regarding specific visual conditions. In this study, we extensively evaluate state-of-the-art parking lot occupancy detection algorithms, compare their prediction quality with the recently emerged vision transformers, and propose a new pipeline based on EfficientNet architecture. Performed computational experiments have demonstrated the performance increase in the case of our model, which was evaluated on 5 different datasets.
☆ On the Generalization of Multi-modal Contrastive Learning
Multi-modal contrastive learning (MMCL) has recently garnered considerable interest due to its superior performance in visual tasks, achieved by embedding multi-modal data, such as visual-language pairs. However, there still lack theoretical understandings of how MMCL extracts useful visual representation from multi-modal pairs, and particularly, how MMCL outperforms previous approaches like self-supervised contrastive learning (SSCL). In this paper, by drawing an intrinsic connection between MMCL and asymmetric matrix factorization, we establish the first generalization guarantees of MMCL for visual downstream tasks. Based on this framework, we further unify MMCL and SSCL by showing that MMCL implicitly performs SSCL with (pseudo) positive pairs induced by text pairs. Through this unified perspective, we characterize the advantage of MMCL by showing that text pairs induce more semantically consistent and diverse positive pairs, which, according to our analysis, provably benefit downstream generalization. Inspired by this finding, we propose CLIP-guided resampling methods to significantly improve the downstream performance of SSCL on ImageNet by leveraging multi-modal information. Code is available at https://github.com/PKU-ML/CLIP-Help-SimCLR.
☆ ColNav: Real-Time Colon Navigation for Colonoscopy
Colorectal cancer screening through colonoscopy continues to be the dominant global standard, as it allows identifying pre-cancerous or adenomatous lesions and provides the ability to remove them during the procedure itself. Nevertheless, failure by the endoscopist to identify such lesions increases the likelihood of lesion progression to subsequent colorectal cancer. Ultimately, colonoscopy remains operator-dependent, and the wide range of quality in colonoscopy examinations among endoscopists is influenced by variations in their technique, training, and diligence. This paper presents a novel real-time navigation guidance system for Optical Colonoscopy (OC). Our proposed system employs a real-time approach that displays both an unfolded representation of the colon and a local indicator directing to un-inspected areas. These visualizations are presented to the physician during the procedure, providing actionable and comprehensible guidance to un-surveyed areas in real-time, while seamlessly integrating into the physician's workflow. Through coverage experimental evaluation, we demonstrated that our system resulted in a higher polyp recall (PR) and high inter-rater reliability with physicians for coverage prediction. These results suggest that our real-time navigation guidance system has the potential to improve the quality and effectiveness of Optical Colonoscopy and ultimately benefit patient outcomes.
☆ DEMIST: A deep-learning-based task-specific denoising approach for myocardial perfusion SPECT
There is an important need for methods to reduce radiation dose and imaging time in myocardial perfusion imaging (MPI) SPECT. Deep learning (DL) methods have demonstrated promise in predicting normal-count images from low-count images for MPI SPECT, but the methods that have been objectively evaluated on the clinical task of detecting perfusion defects have not shown improved performance compared with low-count images. To address this need, we build upon concepts from model-observer theory and our understanding of the human visual system to propose a Detection task-specific DL-based approach for denoising MPI SPECT images (DEMIST). The approach, while performing denoising, is designed to preserve features that are known to impact observer performance on detection tasks. We objectively evaluated the proposed method on the task of detecting perfusion defects using a retrospective study with anonymized clinical data in patients who underwent MPI studies (N = 338). Performance on the task of detecting perfusion defects was quantified with an anthropomorphic channelized Hotelling observer. Images denoised with DEMIST yielded significantly improved detection performance compared to the corresponding low-dose images and images denoised with a commonly used task-agnostic DL-based denoising method. Similar results were observed with stratified analysis based on patient sex and defect type. Additionally, the proposed method significantly improved performance compared to the low-dose images in terms of the task-agnostic metrics of root mean squared error and structural similarity index metric. A mathematical analysis reveals that DEMIST preserves detection-task-specific features while improving the noise properties, thus resulting in improved observer performance. The results provide strong evidence for further clinical evaluation of DEMIST to denoise low-count images in MPI SPECT.
☆ Coarse Is Better? A New Pipeline Towards Self-Supervised Learning with Uncurated Images
Most self-supervised learning (SSL) methods often work on curated datasets where the object-centric assumption holds. This assumption breaks down in uncurated scene images. Existing scene image SSL methods try to find the two views from original scene images that are well matched or dense, which is both complex and computationally heavy. This paper proposes a conceptually different pipeline: first find regions that are coarse objects (with adequate objectness), crop them out as pseudo object-centric images, then any SSL method can be directly applied as in a real object-centric dataset. That is, cropping benefits scene images SSL. A novel cropping strategy is proposed to find coarse object regions in scene images. The proposed pipeline and cropping strategy successfully learn high quality visual representation from uncurated scene datasets like MS-COCO, and the learning does not rely on external object-centric datasets such as ImageNet. Experiments show that our pipeline outperforms existing SSL methods on scene images, and is friendly for non-contrastive SSL methods like MAE. Ablations further verify that the proposed cropping strategy does not rely on pretrained SSL models, too.
☆ T-ADAF: Adaptive Data Augmentation Framework for Image Classification Network based on Tensor T-product Operator
Image classification is one of the most fundamental tasks in Computer Vision. In practical applications, the datasets are usually not as abundant as those in the laboratory and simulation, which is always called as Data Hungry. How to extract the information of data more completely and effectively is very important. Therefore, an Adaptive Data Augmentation Framework based on the tensor T-product Operator is proposed in this paper, to triple one image data to be trained and gain the result from all these three images together with only less than 0.1% increase in the number of parameters. At the same time, this framework serves the functions of column image embedding and global feature intersection, enabling the model to obtain information in not only spatial but frequency domain, and thus improving the prediction accuracy of the model. Numerical experiments have been designed for several models, and the results demonstrate the effectiveness of this adaptive framework. Numerical experiments show that our data augmentation framework can improve the performance of original neural network model by 2%, which provides competitive results to state-of-the-art methods.
☆ Randomized 3D Scene Generation for Generalizable Self-supervised Pre-training
Capturing and labeling real-world 3D data is laborious and time-consuming, which makes it costly to train strong 3D models. To address this issue, previous works generate randomized 3D scenes and pre-train models on generated data. Although the pre-trained models gain promising performance boosts, previous works have two major shortcomings. First, they focus on only one downstream task (i.e., object detection). Second, a fair comparison of generated data is still lacking. In this work, we systematically compare data generation methods using a unified setup. To clarify the generalization of the pre-trained models, we evaluate their performance in multiple tasks (e.g., object detection and semantic segmentation) and with different pre-training methods (e.g., masked autoencoder and contrastive learning). Moreover, we propose a new method to generate 3D scenes with spherical harmonics. It surpasses the previous formula-driven method with a clear margin and achieves on-par results with methods using real-world scans and CAD models.
☆ Flare7K++: Mixing Synthetic and Real Datasets for Nighttime Flare Removal and Beyond
Artificial lights commonly leave strong lens flare artifacts on the images captured at night, degrading both the visual quality and performance of vision algorithms. Existing flare removal approaches mainly focus on removing daytime flares and fail in nighttime cases. Nighttime flare removal is challenging due to the unique luminance and spectrum of artificial lights, as well as the diverse patterns and image degradation of the flares. The scarcity of the nighttime flare removal dataset constraints the research on this crucial task. In this paper, we introduce Flare7K++, the first comprehensive nighttime flare removal dataset, consisting of 962 real-captured flare images (Flare-R) and 7,000 synthetic flares (Flare7K). Compared to Flare7K, Flare7K++ is particularly effective in eliminating complicated degradation around the light source, which is intractable by using synthetic flares alone. Besides, the previous flare removal pipeline relies on the manual threshold and blur kernel settings to extract light sources, which may fail when the light sources are tiny or not overexposed. To address this issue, we additionally provide the annotations of light sources in Flare7K++ and propose a new end-to-end pipeline to preserve the light source while removing lens flares. Our dataset and pipeline offer a valuable foundation and benchmark for future investigations into nighttime flare removal studies. Extensive experiments demonstrate that Flare7K++ supplements the diversity of existing flare datasets and pushes the frontier of nighttime flare removal towards real-world scenarios.
comment: Extension of arXiv:2210.06570; Project page at https://ykdai.github.io/projects/Flare7K
☆ Learning Probabilistic Coordinate Fields for Robust Correspondences
We introduce Probabilistic Coordinate Fields (PCFs), a novel geometric-invariant coordinate representation for image correspondence problems. In contrast to standard Cartesian coordinates, PCFs encode coordinates in correspondence-specific barycentric coordinate systems (BCS) with affine invariance. To know \textit{when and where to trust} the encoded coordinates, we implement PCFs in a probabilistic network termed PCF-Net, which parameterizes the distribution of coordinate fields as Gaussian mixture models. By jointly optimizing coordinate fields and their confidence conditioned on dense flows, PCF-Net can work with various feature descriptors when quantifying the reliability of PCFs by confidence maps. An interesting observation of this work is that the learned confidence map converges to geometrically coherent and semantically consistent regions, which facilitates robust coordinate representation. By delivering the confident coordinates to keypoint/feature descriptors, we show that PCF-Net can be used as a plug-in to existing correspondence-dependent approaches. Extensive experiments on both indoor and outdoor datasets suggest that accurate geometric invariant coordinates help to achieve the state of the art in several correspondence problems, such as sparse feature matching, dense image registration, camera pose estimation, and consistency filtering. Further, the interpretable confidence map predicted by PCF-Net can also be leveraged to other novel applications from texture transfer to multi-homography classification.
comment: Accepted by IEEE Transactions on Pattern Analysis and Machine Intelligence
☆ Normalization Layers Are All That Sharpness-Aware Minimization Needs
Sharpness-aware minimization (SAM) was proposed to reduce sharpness of minima and has been shown to enhance generalization performance in various settings. In this work we show that perturbing only the affine normalization parameters (comprising less than 0.1% of the total parameters) in the adversarial step of SAM outperforms perturbing all of the parameters. This finding generalizes to different SAM variants and both ResNet (Batch Normalization) and Vision Transformer (Layer Normalization) architectures. We consider alternative sparse perturbation approaches and find that these do not achieve similar performance enhancement at such extreme sparsity levels, showing that this behaviour is unique to the normalization layers. Although our findings reaffirm the effectiveness of SAM in improving generalization performance, they cast doubt on whether this is solely caused by reduced sharpness. The code for our experiments is publicly available at https://github.com/mueller-mp/SAM-ON.
☆ Efficient Vision Transformer for Human Pose Estimation via Patch Selection
While Convolutional Neural Networks (CNNs) have been widely successful in 2D human pose estimation, Vision Transformers (ViTs) have emerged as a promising alternative to CNNs, boosting state-of-the-art performance. However, the quadratic computational complexity of ViTs has limited their applicability for processing high-resolution images and long videos. To address this challenge, we propose a simple method for reducing ViT's computational complexity based on selecting and processing a small number of most informative patches while disregarding others. We leverage a lightweight pose estimation network to guide the patch selection process, ensuring that the selected patches contain the most important information. Our experimental results on three widely used 2D pose estimation benchmarks, namely COCO, MPII and OCHuman, demonstrate the effectiveness of our proposed methods in significantly improving speed and reducing computational complexity with a slight drop in performance.
☆ MultiSum: A Dataset for Multimodal Summarization and Thumbnail Generation of Videos
Multimodal summarization with multimodal output (MSMO) has emerged as a promising research direction. Nonetheless, numerous limitations exist within existing public MSMO datasets, including insufficient upkeep, data inaccessibility, limited size, and the absence of proper categorization, which pose significant challenges to effective research. To address these challenges and provide a comprehensive dataset for this new direction, we have meticulously curated the MultiSum dataset. Our new dataset features (1) Human-validated summaries for both video and textual content, providing superior human instruction and labels for multimodal learning. (2) Comprehensively and meticulously arranged categorization, spanning 17 principal categories and 170 subcategories to encapsulate a diverse array of real-world scenarios. (3) Benchmark tests performed on the proposed dataset to assess varied tasks and methods, including video temporal segmentation, video summarization, text summarization, and multimodal summarization. To champion accessibility and collaboration, we release the MultiSum dataset and the data collection tool as fully open-source resources, fostering transparency and accelerating future developments. Our project website can be found at https://multisum-dataset.github.io/.
comment: Project website: https://multisum-dataset.github.io/
☆ DualHGNN: A Dual Hypergraph Neural Network for Semi-Supervised Node Classification based on Multi-View Learning and Density Awareness IJCNN 2023
Graph-based semi-supervised node classification has been shown to become a state-of-the-art approach in many applications with high research value and significance. Most existing methods are only based on the original intrinsic or artificially established graph structure which may not accurately reflect the "true" correlation among data and are not optimal for semi-supervised node classification in the downstream graph neural networks. Besides, while existing graph-based methods mostly utilize the explicit graph structure, some implicit information, for example, the density information, can also provide latent information that can be further exploited. To address these limitations, this paper proposes the Dual Hypergraph Neural Network (DualHGNN), a new dual connection model integrating both hypergraph structure learning and hypergraph representation learning simultaneously in a unified architecture. The DualHGNN first leverages a multi-view hypergraph learning network to explore the optimal hypergraph structure from multiple views, constrained by a consistency loss proposed to improve its generalization. Then, DualHGNN employs a density-aware hypergraph attention network to explore the high-order semantic correlation among data points based on the density-aware attention mechanism. Extensive experiments are conducted in various benchmark datasets, and the results demonstrate the effectiveness of the proposed approach.
comment: This work has been accepted by 2023 International Joint Conference on Neural Networks (IJCNN 2023). arXiv admin note: text overlap with arXiv:2201.11511
☆ Video Compression with Arbitrary Rescaling Network
Most video platforms provide video streaming services with different qualities, and the quality of the services is usually adjusted by the resolution of the videos. So high-resolution videos need to be downsampled for compression. In order to solve the problem of video coding at different resolutions, we propose a rate-guided arbitrary rescaling network (RARN) for video resizing before encoding. To help the RARN be compatible with standard codecs and generate compression-friendly results, an iteratively optimized transformer-based virtual codec (TVC) is introduced to simulate the key components of video encoding and perform bitrate estimation. By iteratively training the TVC and the RARN, we achieved 5%-29% BD-Rate reduction anchored by linear interpolation under different encoding configurations and resolutions, exceeding the previous methods on most test videos. Furthermore, the lightweight RARN structure can process FHD (1080p) content at real-time speed (91 FPS) and obtain a considerable rate reduction.
comment: Accepted as a one-page poster by 2023 Data Compression Conference (DCC). This is the full paper
☆ StructuredMesh: 3D Structured Optimization of Façade Components on Photogrammetric Mesh Models using Binary Integer Programming
The lack of fa\c{c}ade structures in photogrammetric mesh models renders them inadequate for meeting the demands of intricate applications. Moreover, these mesh models exhibit irregular surfaces with considerable geometric noise and texture quality imperfections, making the restoration of structures challenging. To address these shortcomings, we present StructuredMesh, a novel approach for reconstructing fa\c{c}ade structures conforming to the regularity of buildings within photogrammetric mesh models. Our method involves capturing multi-view color and depth images of the building model using a virtual camera and employing a deep learning object detection pipeline to semi-automatically extract the bounding boxes of fa\c{c}ade components such as windows, doors, and balconies from the color image. We then utilize the depth image to remap these boxes into 3D space, generating an initial fa\c{c}ade layout. Leveraging architectural knowledge, we apply binary integer programming (BIP) to optimize the 3D layout's structure, encompassing the positions, orientations, and sizes of all components. The refined layout subsequently informs fa\c{c}ade modeling through instance replacement. We conducted experiments utilizing building mesh models from three distinct datasets, demonstrating the adaptability, robustness, and noise resistance of our proposed methodology. Furthermore, our 3D layout evaluation metrics reveal that the optimized layout enhances precision, recall, and F-score by 6.5%, 4.5%, and 5.5%, respectively, in comparison to the initial layout.
comment: 30 pages,15 figures
☆ FusedRF: Fusing Multiple Radiance Fields CVPR
Radiance Fields (RFs) have shown great potential to represent scenes from casually captured discrete views. Compositing parts or whole of multiple captured scenes could greatly interest several XR applications. Prior works can generate new views of such scenes by tracing each scene in parallel. This increases the render times and memory requirements with the number of components. In this work, we provide a method to create a single, compact, fused RF representation for a scene composited using multiple RFs. The fused RF has the same render times and memory utilizations as a single RF. Our method distills information from multiple teacher RFs into a single student RF while also facilitating further manipulations like addition and deletion into the fused representation.
comment: XRNeRF CVPR Workshop Paper
☆ ScoreCL: Augmentation-Adaptive Contrastive Learning via Score-Matching Function
Self-supervised contrastive learning (CL) has achieved state-of-the-art performance in representation learning by minimizing the distance between positive pairs while maximizing that of negative ones. Recently, it has been verified that the model learns better representation with diversely augmented positive pairs because they enable the model to be more view-invariant. However, only a few studies on CL have considered the difference between augmented views, and have not gone beyond the hand-crafted findings. In this paper, we first observe that the score-matching function can measure how much data has changed from the original through augmentation. With the observed property, every pair in CL can be weighted adaptively by the difference of score values, resulting in boosting the performance of the existing CL method. We show the generality of our method, referred to as ScoreCL, by consistently improving various CL methods, SimCLR, SimSiam, W-MSE, and VICReg, up to 3%p in k-NN evaluation on CIFAR-10, CIFAR-100, and ImageNet-100. Moreover, we have conducted exhaustive experiments and ablations, including results on diverse downstream tasks, comparison with possible baselines, and improvement when used with other proposed augmentation methods. We hope our exploration will inspire more research in exploiting the score matching for CL.
☆ BAA-NGP: Bundle-Adjusting Accelerated Neural Graphics Primitives
Implicit neural representation has emerged as a powerful method for reconstructing 3D scenes from 2D images. Given a set of camera poses and associated images, the models can be trained to synthesize novel, unseen views. In order to expand the use cases for implicit neural representations, we need to incorporate camera pose estimation capabilities as part of the representation learning, as this is necessary for reconstructing scenes from real-world video sequences where cameras are generally not being tracked. Existing approaches like COLMAP and, most recently, bundle-adjusting neural radiance field methods often suffer from lengthy processing times. These delays ranging from hours to days, arise from laborious feature matching, hardware limitations, dense point sampling, and long training times required by a multi-layer perceptron structure with a large number of parameters. To address these challenges, we propose a framework called bundle-adjusting accelerated neural graphics primitives (BAA-NGP). Our approach leverages accelerated sampling and hash encoding to expedite both pose refinement/estimation and 3D scene reconstruction. Experimental results demonstrate that our method achieves a more than 10 to 20 $\times$ speed improvement in novel view synthesis compared to other bundle-adjusting neural radiance field methods without sacrificing the quality of pose estimation.
☆ Enhancing Virtual Assistant Intelligence: Precise Area Targeting for Instance-level User Intents beyond Metadata
Virtual assistants have been widely used by mobile phone users in recent years. Although their capabilities of processing user intents have been developed rapidly, virtual assistants in most platforms are only capable of handling pre-defined high-level tasks supported by extra manual efforts of developers. However, instance-level user intents containing more detailed objectives with complex practical situations, are yet rarely studied so far. In this paper, we explore virtual assistants capable of processing instance-level user intents based on pixels of application screens, without the requirements of extra extensions on the application side. We propose a novel cross-modal deep learning pipeline, which understands the input vocal or textual instance-level user intents, predicts the targeting operational area, and detects the absolute button area on screens without any metadata of applications. We conducted a user study with 10 participants to collect a testing dataset with instance-level user intents. The testing dataset is then utilized to evaluate the performance of our model, which demonstrates that our model is promising with the achievement of 64.43% accuracy on our testing dataset.
☆ CFDP: Common Frequency Domain Pruning CVPR
As the saying goes, sometimes less is more -- and when it comes to neural networks, that couldn't be more true. Enter pruning, the art of selectively trimming away unnecessary parts of a network to create a more streamlined, efficient architecture. In this paper, we introduce a novel end-to-end pipeline for model pruning via the frequency domain. This work aims to shed light on the interoperability of intermediate model outputs and their significance beyond the spatial domain. Our method, dubbed Common Frequency Domain Pruning (CFDP) aims to extrapolate common frequency characteristics defined over the feature maps to rank the individual channels of a layer based on their level of importance in learning the representation. By harnessing the power of CFDP, we have achieved state-of-the-art results on CIFAR-10 with GoogLeNet reaching an accuracy of 95.25%, that is, +0.2% from the original model. We also outperform all benchmarks and match the original model's performance on ImageNet, using only 55% of the trainable parameters and 60% of the FLOPs. In addition to notable performances, models produced via CFDP exhibit robustness to a variety of configurations including pruning from untrained neural architectures, and resistance to adversarial attacks. The implementation code can be found at https://github.com/Skhaki18/CFDP.
comment: CVPR ECV 2023 Accepted Paper
☆ Matte Anything: Interactive Natural Image Matting with Segment Anything Models
Natural image matting algorithms aim to predict the transparency map (alpha-matte) with the trimap guidance. However, the production of trimaps often requires significant labor, which limits the widespread application of matting algorithms on a large scale. To address the issue, we propose Matte Anything model (MatAny), an interactive natural image matting model which could produce high-quality alpha-matte with various simple hints. The key insight of MatAny is to generate pseudo trimap automatically with contour and transparency prediction. We leverage task-specific vision models to enhance the performance of natural image matting. Specifically, we use the segment anything model (SAM) to predict high-quality contour with user interaction and an open-vocabulary (OV) detector to predict the transparency of any object. Subsequently, a pretrained image matting model generates alpha mattes with pseudo trimaps. MatAny is the interactive matting algorithm with the most supported interaction methods and the best performance to date. It consists of orthogonal vision models without any additional training. We evaluate the performance of MatAny against several current image matting algorithms, and the results demonstrate the significant potential of our approach.
comment: Technical Report
☆ Manga Rescreening with Interpretable Screentone Representation
The process of adapting or repurposing manga pages is a time-consuming task that requires manga artists to manually work on every single screentone region and apply new patterns to create novel screentones across multiple panels. To address this issue, we propose an automatic manga rescreening pipeline that aims to minimize the human effort involved in manga adaptation. Our pipeline automatically recognizes screentone regions and generates novel screentones with newly specified characteristics (e.g., intensity or type). Existing manga generation methods have limitations in understanding and synthesizing complex tone- or intensity-varying regions. To overcome these limitations, we propose a novel interpretable representation of screentones that disentangles their intensity and type features, enabling better recognition and synthesis of screentones. This interpretable screentone representation reduces ambiguity in recognizing intensity-varying regions and provides fine-grained controls during screentone synthesis by decoupling and anchoring the type or the intensity feature. Our proposed method is demonstrated to be effective and convenient through various experiments, showcasing the superiority of the newly proposed pipeline with the interpretable screentone representations.
comment: 10 pages, 11 figures
☆ Phoenix: A Federated Generative Diffusion Model
Generative AI has made impressive strides in enabling users to create diverse and realistic visual content such as images, videos, and audio. However, training generative models on large centralized datasets can pose challenges in terms of data privacy, security, and accessibility. Federated learning (FL) is an approach that uses decentralized techniques to collaboratively train a shared deep learning model while retaining the training data on individual edge devices to preserve data privacy. This paper proposes a novel method for training a Denoising Diffusion Probabilistic Model (DDPM) across multiple data sources using FL techniques. Diffusion models, a newly emerging generative model, show promising results in achieving superior quality images than Generative Adversarial Networks (GANs). Our proposed method Phoenix is an unconditional diffusion model that leverages strategies to improve the data diversity of generated samples even when trained on data with statistical heterogeneity or Non-IID (Non-Independent and Identically Distributed) data. We demonstrate how our approach outperforms the default diffusion model in an FL setting. These results indicate that high-quality samples can be generated by maintaining data diversity, preserving privacy, and reducing communication between data sources, offering exciting new possibilities in the field of generative AI.
☆ 1st Place Solution for PVUW Challenge 2023: Video Panoptic Segmentation
Video panoptic segmentation is a challenging task that serves as the cornerstone of numerous downstream applications, including video editing and autonomous driving. We believe that the decoupling strategy proposed by DVIS enables more effective utilization of temporal information for both "thing" and "stuff" objects. In this report, we successfully validated the effectiveness of the decoupling strategy in video panoptic segmentation. Finally, our method achieved a VPQ score of 51.4 and 53.7 in the development and test phases, respectively, and ultimately ranked 1st in the VPS track of the 2nd PVUW Challenge. The code is available at https://github.com/zhang-tao-whu/DVIS
☆ TEC-Net: Vision Transformer Embrace Convolutional Neural Networks for Medical Image Segmentation
The hybrid architecture of convolution neural networks (CNN) and Transformer has been the most popular method for medical image segmentation. However, the existing networks based on the hybrid architecture suffer from two problems. First, although the CNN branch can capture image local features by using convolution operation, the vanilla convolution is unable to achieve adaptive extraction of image features. Second, although the Transformer branch can model the global information of images, the conventional self-attention only focuses on the spatial self-attention of images and ignores the channel and cross-dimensional self-attention leading to low segmentation accuracy for medical images with complex backgrounds. To solve these problems, we propose vision Transformer embrace convolutional neural networks for medical image segmentation (TEC-Net). Our network has two advantages. First, dynamic deformable convolution (DDConv) is designed in the CNN branch, which not only overcomes the difficulty of adaptive feature extraction using fixed-size convolution kernels, but also solves the defect that different inputs share the same convolution kernel parameters, effectively improving the feature expression ability of CNN branch. Second, in the Transformer branch, a (shifted)-window adaptive complementary attention module ((S)W-ACAM) and compact convolutional projection are designed to enable the network to fully learn the cross-dimensional long-range dependency of medical images with few parameters and calculations. Experimental results show that the proposed TEC-Net provides better medical image segmentation results than SOTA methods including CNN and Transformer networks. In addition, our TEC-Net requires fewer parameters and computational costs and does not rely on pre-training. The code is publicly available at https://github.com/SR0920/TEC-Net.
comment: arXiv admin note: substantial text overlap with arXiv:2306.03373
♻ ☆ Balanced Product of Calibrated Experts for Long-Tailed Recognition CVPR 2023
Many real-world recognition problems are characterized by long-tailed label distributions. These distributions make representation learning highly challenging due to limited generalization over the tail classes. If the test distribution differs from the training distribution, e.g. uniform versus long-tailed, the problem of the distribution shift needs to be addressed. A recent line of work proposes learning multiple diverse experts to tackle this issue. Ensemble diversity is encouraged by various techniques, e.g. by specializing different experts in the head and the tail classes. In this work, we take an analytical approach and extend the notion of logit adjustment to ensembles to form a Balanced Product of Experts (BalPoE). BalPoE combines a family of experts with different test-time target distributions, generalizing several previous approaches. We show how to properly define these distributions and combine the experts in order to achieve unbiased predictions, by proving that the ensemble is Fisher-consistent for minimizing the balanced error. Our theoretical analysis shows that our balanced ensemble requires calibrated experts, which we achieve in practice using mixup. We conduct extensive experiments and our method obtains new state-of-the-art results on three long-tailed datasets: CIFAR-100-LT, ImageNet-LT, and iNaturalist-2018. Our code is available at https://github.com/emasa/BalPoE-CalibratedLT.
comment: Accepted at CVPR 2023, 19 pages
♻ ☆ Progressive Update Guided Interdependent Networks for Single Image Dehazing
Images with haze of different varieties often pose a significant challenge to dehazing. Therefore, guidance by estimates of haze parameters related to the variety would be beneficial, and their progressive update jointly with haze reduction will allow effective dehazing. To this end, we propose a multi-network dehazing framework containing novel interdependent dehazing and haze parameter updater networks that operate in a progressive manner. The haze parameters, transmission map and atmospheric light, are first estimated using dedicated convolutional networks that allow color-cast handling. The estimated parameters are then used to guide our dehazing module, where the estimates are progressively updated by novel convolutional networks. The updating takes place jointly with progressive dehazing using a network that invokes inter-step dependencies. The joint progressive updating and dehazing gradually modify the haze parameter values toward achieving effective dehazing. Through different studies, our dehazing framework is shown to be more effective than image-to-image mapping and predefined haze formation model based dehazing. The framework is also found capable of handling a wide variety of hazy conditions wtih different types and amounts of haze and color casts. Our dehazing framework is qualitatively and quantitatively found to outperform the state-of-the-art on synthetic and real-world hazy images of multiple datasets with varied haze conditions.
comment: First two authors contributed equally. This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible. Project Website: https://aupendu.github.io/progressive-dehaze
♻ ☆ Mixed Autoencoder for Self-supervised Visual Representation Learning CVPR 2023
Masked Autoencoder (MAE) has demonstrated superior performance on various vision tasks via randomly masking image patches and reconstruction. However, effective data augmentation strategies for MAE still remain open questions, different from those in contrastive learning that serve as the most important part. This paper studies the prevailing mixing augmentation for MAE. We first demonstrate that naive mixing will in contrast degenerate model performance due to the increase of mutual information (MI). To address, we propose homologous recognition, an auxiliary pretext task, not only to alleviate the MI increasement by explicitly requiring each patch to recognize homologous patches, but also to perform object-aware self-supervised pre-training for better downstream dense perception performance. With extensive experiments, we demonstrate that our proposed Mixed Autoencoder (MixedAE) achieves the state-of-the-art transfer results among masked image modeling (MIM) augmentations on different downstream tasks with significant efficiency. Specifically, our MixedAE outperforms MAE by +0.3% accuracy, +1.7 mIoU and +0.9 AP on ImageNet-1K, ADE20K and COCO respectively with a standard ViT-Base. Moreover, MixedAE surpasses iBOT, a strong MIM method combined with instance discrimination, while accelerating training by 2x. To our best knowledge, this is the very first work to consider mixing for MIM from the perspective of pretext task design. Code will be made available.
comment: Accepted by CVPR 2023
♻ ☆ Easily Accessible Text-to-Image Generation Amplifies Demographic Stereotypes at Large Scale
Machine learning models that convert user-written text descriptions into images are now widely available online and used by millions of users to generate millions of images a day. We investigate the potential for these models to amplify dangerous and complex stereotypes. We find a broad range of ordinary prompts produce stereotypes, including prompts simply mentioning traits, descriptors, occupations, or objects. For example, we find cases of prompting for basic traits or social roles resulting in images reinforcing whiteness as ideal, prompting for occupations resulting in amplification of racial and gender disparities, and prompting for objects resulting in reification of American norms. Stereotypes are present regardless of whether prompts explicitly mention identity and demographic language or avoid such language. Moreover, stereotypes persist despite mitigation strategies; neither user attempts to counter stereotypes by requesting images with specific counter-stereotypes nor institutional attempts to add system ``guardrails'' have prevented the perpetuation of stereotypes. Our analysis justifies concerns regarding the impacts of today's models, presenting striking exemplars, and connecting these findings with deep insights into harms drawn from social scientific and humanist disciplines. This work contributes to the effort to shed light on the uniquely complex biases in language-vision models and demonstrates the ways that the mass deployment of text-to-image generation models results in mass dissemination of stereotypes and resulting harms.
comment: FAccT 2023 paper. The published version is available at 10.1145/3593013.3594095
♻ ☆ Geo-Tiles for Semantic Segmentation of Earth Observation Imagery
To cope with the high requirements during the computation of semantic segmentations of earth observation imagery, current state-of-the-art pipelines divide the corresponding data into smaller images. Existing methods and benchmark datasets oftentimes rely on pixel-based tiling schemes or on geo-tiling schemes employed by web mapping applications. The selection of subimages (comprising size, location and orientation) is crucial. It affects the available context information of each pixel, defines the number of tiles during training, and influences the degree of information degradation while down- and up-sampling the tile contents to the size required by the segmentation model. We propose a new segmentation pipeline for earth observation imagery relying on a tiling scheme that creates geo-tiles based on the geo-information of the raster data. This approach exhibits several beneficial properties compared to pixel-based or common web mapping approaches. The proposed tiling scheme shows flexible customization properties regarding tile granularity, tile stride and image boundary alignment. This allows us to perform a tile specific data augmentation during training and a substitution of pixel predictions with limited context information using data of overlapping tiles during inference. The generated tiles show a consistent spatial tile extent w.r.t. heterogeneous sensors, varying recording distances and different latitudes. We demonstrate how the proposed tiling system allows to improve the results of current state-of-the-art semantic segmentation models. To foster future research we make the source code publicly available.
♻ ☆ Self-supervised learning-based cervical cytology for the triage of HPV-positive women in resource-limited settings and low-data regime
Screening Papanicolaou test samples has proven to be highly effective in reducing cervical cancer-related mortality. However, the lack of trained cytopathologists hinders its widespread implementation in low-resource settings. Deep learning-based telecytology diagnosis emerges as an appealing alternative, but it requires the collection of large annotated training datasets, which is costly and time-consuming. In this paper, we demonstrate that the abundance of unlabeled images that can be extracted from Pap smear test whole slide images presents a fertile ground for self-supervised learning methods, yielding performance improvements relative to readily available pre-trained models for various downstream tasks. In particular, we propose \textbf{C}ervical \textbf{C}ell \textbf{C}opy-\textbf{P}asting ($\texttt{C}^{3}\texttt{P}$) as an effective augmentation method, which enables knowledge transfer from open-source and labeled single-cell datasets to unlabeled tiles. Not only does $\texttt{C}^{3}\texttt{P}$ outperforms naive transfer from single-cell images, but we also demonstrate its advantageous integration into multiple instance learning methods. Importantly, all our experiments are conducted on our introduced \textit{in-house} dataset comprising liquid-based cytology Pap smear images obtained using low-cost technologies. This aligns with our objective of leveraging deep learning-based telecytology for diagnosis in low-resource settings.
♻ ☆ Efficient Video Action Detection with Token Dropout and Context Refinement
Streaming video clips with large-scale video tokens impede vision transformers (ViTs) for efficient recognition, especially in video action detection where sufficient spatiotemporal representations are required for precise actor identification. In this work, we propose an end-to-end framework for efficient video action detection (EVAD) based on vanilla ViTs. Our EVAD consists of two specialized designs for video action detection. First, we propose a spatiotemporal token dropout from a keyframe-centric perspective. In a video clip, we maintain all tokens from its keyframe, preserve tokens relevant to actor motions from other frames, and drop out the remaining tokens in this clip. Second, we refine scene context by leveraging remaining tokens for better recognizing actor identities. The region of interest (RoI) in our action detector is expanded into temporal domain. The captured spatiotemporal actor identity representations are refined via scene context in a decoder with the attention mechanism. These two designs make our EVAD efficient while maintaining accuracy, which is validated on three benchmark datasets (i.e., AVA, UCF101-24, JHMDB). Compared to the vanilla ViT backbone, our EVAD reduces the overall GFLOPs by 43% and improves real-time inference speed by 40% with no performance degradation. Moreover, even at similar computational costs, our EVAD can improve the performance by 1.1 mAP with higher resolution inputs. Code is available at https://github.com/MCG-NJU/EVAD.
comment: technical report
♻ ☆ Time-Conditioned Generative Modeling of Object-Centric Representations for Video Decomposition and Prediction
When perceiving the world from multiple viewpoints, humans have the ability to reason about the complete objects in a compositional manner even when an object is completely occluded from certain viewpoints. Meanwhile, humans are able to imagine novel views after observing multiple viewpoints. Recent remarkable advances in multi-view object-centric learning still leaves some unresolved problems: 1) The shapes of partially or completely occluded objects can not be well reconstructed. 2) The novel viewpoint prediction depends on expensive viewpoint annotations rather than implicit rules in view representations. In this paper, we introduce a time-conditioned generative model for videos. To reconstruct the complete shape of an object accurately, we enhance the disentanglement between the latent representations of objects and views, where the latent representations of time-conditioned views are jointly inferred with a Transformer and then are input to a sequential extension of Slot Attention to learn object-centric representations. In addition, Gaussian processes are employed as priors of view latent variables for video generation and novel-view prediction without viewpoint annotations. Experiments on multiple datasets demonstrate that the proposed model can make object-centric video decomposition, reconstruct the complete shapes of occluded objects, and make novel-view predictions.
♻ ☆ FPUS23: An Ultrasound Fetus Phantom Dataset with Deep Neural Network Evaluations for Fetus Orientations, Fetal Planes, and Anatomical Features
Ultrasound imaging is one of the most prominent technologies to evaluate the growth, progression, and overall health of a fetus during its gestation. However, the interpretation of the data obtained from such studies is best left to expert physicians and technicians who are trained and well-versed in analyzing such images. To improve the clinical workflow and potentially develop an at-home ultrasound-based fetal monitoring platform, we present a novel fetus phantom ultrasound dataset, FPUS23, which can be used to identify (1) the correct diagnostic planes for estimating fetal biometric values, (2) fetus orientation, (3) their anatomical features, and (4) bounding boxes of the fetus phantom anatomies at 23 weeks gestation. The entire dataset is composed of 15,728 images, which are used to train four different Deep Neural Network models, built upon a ResNet34 backbone, for detecting aforementioned fetus features and use-cases. We have also evaluated the models trained using our FPUS23 dataset, to show that the information learned by these models can be used to substantially increase the accuracy on real-world ultrasound fetus datasets. We make the FPUS23 dataset and the pre-trained models publicly accessible at https://github.com/bharathprabakaran/FPUS23, which will further facilitate future research on fetal ultrasound imaging and analysis.
comment: Accepted for Publication at IEEE Access
♻ ☆ Introduction to Medical Imaging Informatics
Medical imaging informatics is a rapidly growing field that combines the principles of medical imaging and informatics to improve the acquisition, management, and interpretation of medical images. This chapter introduces the basic concepts of medical imaging informatics, including image processing, feature engineering, and machine learning. It also discusses the recent advancements in computer vision and deep learning technologies and how they are used to develop new quantitative image markers and prediction models for disease detection, diagnosis, and prognosis prediction. By covering the basic knowledge of medical imaging informatics, this chapter provides a foundation for understanding the role of informatics in medicine and its potential impact on patient care.
comment: 18 pages, 11 figures, 2 tables; Acceptance of the chapter for the Springer book "Data-driven approaches to medical imaging"
♻ ☆ GPT4Image: Can Large Pre-trained Models Help Vision Models on Perception Tasks?
The recent upsurge in pre-trained large models (e.g. GPT-4) has swept across the entire deep learning community. Such powerful large language models (LLMs) demonstrate advanced generative ability and multimodal understanding capability, which quickly achieve new state-of-the-art performances on a variety of benchmarks. The pre-trained LLM usually plays the role as a universal AI model that can conduct various tasks, including context reasoning, article analysis and image content comprehension. However, considering the prohibitively high memory and computational cost for implementing such a large model, the conventional models (such as CNN and ViT), are still essential for many visual perception tasks. In this paper, we propose to enhance the representation ability of ordinary vision models for perception tasks (e.g. image classification) by taking advantage of large pre-trained models. We present a new learning paradigm in which the knowledge extracted from large pre-trained models are utilized to help models like CNN and ViT learn enhanced representations and achieve better performance. Firstly, we curate a high quality description set by prompting a multimodal LLM to generate descriptive text for all training images. Furthermore, we feed these detailed descriptions into a pre-trained encoder to extract text embeddings with rich semantic information that encodes the content of images. During training, text embeddings will serve as extra supervising signals and be aligned with image representations learned by vision models. The alignment process helps vision models learn better and achieve higher accuracy with the assistance of pre-trained LLMs. We conduct extensive experiments to verify that the proposed algorithm consistently improves the performance for various vision models with heterogeneous architectures.
comment: GitHub: https://github.com/huawei-noah/Efficient-Computing/tree/master/GPT4Image/
♻ ☆ Open-TransMind: A New Baseline and Benchmark for 1st Foundation Model Challenge of Intelligent Transportation
With the continuous improvement of computing power and deep learning algorithms in recent years, the foundation model has grown in popularity. Because of its powerful capabilities and excellent performance, this technology is being adopted and applied by an increasing number of industries. In the intelligent transportation industry, artificial intelligence faces the following typical challenges: few shots, poor generalization, and a lack of multi-modal techniques. Foundation model technology can significantly alleviate the aforementioned issues. To address these, we designed the 1st Foundation Model Challenge, with the goal of increasing the popularity of foundation model technology in traffic scenarios and promoting the rapid development of the intelligent transportation industry. The challenge is divided into two tracks: all-in-one and cross-modal image retrieval. Furthermore, we provide a new baseline and benchmark for the two tracks, called Open-TransMind. According to our knowledge, Open-TransMind is the first open-source transportation foundation model with multi-task and multi-modal capabilities. Simultaneously, Open-TransMind can achieve state-of-the-art performance on detection, classification, and segmentation datasets of traffic scenarios. Our source code is available at https://github.com/Traffic-X/Open-TransMind.
♻ ☆ MixPHM: Redundancy-Aware Parameter-Efficient Tuning for Low-Resource Visual Question Answering CVPR 2023
Recently, finetuning pretrained Vision-Language Models (VLMs) has been a prevailing paradigm for achieving state-of-the-art performance in Visual Question Answering (VQA). However, as VLMs scale, finetuning full model parameters for a given task in low-resource settings becomes computationally expensive, storage inefficient, and prone to overfitting. Current parameter-efficient tuning methods dramatically reduce the number of tunable parameters, but there still exists a significant performance gap with full finetuning. In this paper, we propose MixPHM, a redundancy-aware parameter-efficient tuning method that outperforms full finetuning in low-resource VQA. Specifically, MixPHM is a lightweight module implemented by multiple PHM-experts in a mixture-of-experts manner. To reduce parameter redundancy, MixPHM reparameterizes expert weights in a low-rank subspace and shares part of the weights inside and across experts. Moreover, based on a quantitative redundancy analysis for adapters, we propose Redundancy Regularization to reduce task-irrelevant redundancy while promoting task-relevant correlation in MixPHM representations. Experiments conducted on VQA v2, GQA, and OK-VQA demonstrate that MixPHM outperforms state-of-the-art parameter-efficient methods and is the only one consistently surpassing full finetuning.
comment: 13 pages, 6 figures, 9 tables. Accepted to CVPR 2023
♻ ☆ Multi-site, Multi-domain Airway Tree Modeling
Open international challenges are becoming the de facto standard for assessing computer vision and image analysis algorithms. In recent years, new methods have extended the reach of pulmonary airway segmentation that is closer to the limit of image resolution. Since EXACT'09 pulmonary airway segmentation, limited effort has been directed to quantitative comparison of newly emerged algorithms driven by the maturity of deep learning based approaches and clinical drive for resolving finer details of distal airways for early intervention of pulmonary diseases. Thus far, public annotated datasets are extremely limited, hindering the development of data-driven methods and detailed performance evaluation of new algorithms. To provide a benchmark for the medical imaging community, we organized the Multi-site, Multi-domain Airway Tree Modeling (ATM'22), which was held as an official challenge event during the MICCAI 2022 conference. ATM'22 provides large-scale CT scans with detailed pulmonary airway annotation, including 500 CT scans (300 for training, 50 for validation, and 150 for testing). The dataset was collected from different sites and it further included a portion of noisy COVID-19 CTs with ground-glass opacity and consolidation. Twenty-three teams participated in the entire phase of the challenge and the algorithms for the top ten teams are reviewed in this paper. Quantitative and qualitative results revealed that deep learning models embedded with the topological continuity enhancement achieved superior performance in general. ATM'22 challenge holds as an open-call design, the training data and the gold standard evaluation are available upon successful registration via its homepage.
comment: 32 pages, 16 figures. Homepage: https://atm22.grand-challenge.org/. Submitted
♻ ☆ Preserving Privacy in Surgical Video Analysis Using Artificial Intelligence: A Deep Learning Classifier to Identify Out-of-Body Scenes in Endoscopic Videos
Objective: To develop and validate a deep learning model for the identification of out-of-body images in endoscopic videos. Background: Surgical video analysis facilitates education and research. However, video recordings of endoscopic surgeries can contain privacy-sensitive information, especially if out-of-body scenes are recorded. Therefore, identification of out-of-body scenes in endoscopic videos is of major importance to preserve the privacy of patients and operating room staff. Methods: A deep learning model was trained and evaluated on an internal dataset of 12 different types of laparoscopic and robotic surgeries. External validation was performed on two independent multicentric test datasets of laparoscopic gastric bypass and cholecystectomy surgeries. All images extracted from the video datasets were annotated as inside or out-of-body. Model performance was evaluated compared to human ground truth annotations measuring the receiver operating characteristic area under the curve (ROC AUC). Results: The internal dataset consisting of 356,267 images from 48 videos and the two multicentric test datasets consisting of 54,385 and 58,349 images from 10 and 20 videos, respectively, were annotated. Compared to ground truth annotations, the model identified out-of-body images with 99.97% ROC AUC on the internal test dataset. Mean $\pm$ standard deviation ROC AUC on the multicentric gastric bypass dataset was 99.94$\pm$0.07% and 99.71$\pm$0.40% on the multicentric cholecystectomy dataset, respectively. Conclusion: The proposed deep learning model can reliably identify out-of-body images in endoscopic videos. The trained model is publicly shared. This facilitates privacy preservation in surgical video analysis.
comment: Jo\"el L. Lavanchy and Armine Vardazaryan contributed equally and share first co-authorship
♻ ☆ CLIP-VG: Self-paced Curriculum Adapting of CLIP for Visual Grounding
Visual Grounding (VG) is a crucial topic in the field of vision and language, which involves locating a specific region described by expressions within an image. To reduce the reliance on manually labeled data, unsupervised methods have been developed to locate regions using pseudo-labels. However, the performance of existing unsupervised methods is highly dependent on the quality of pseudo-labels and these methods always encounter issues with limited diversity. In order to utilize vision and language pre-trained models to address the grounding problem, and reasonably take advantage of pseudo-labels, we propose CLIP-VG, a novel method that can conduct self-paced curriculum adapting of CLIP with pseudo-language labels. We propose a simple yet efficient end-to-end network architecture to realize the transfer of CLIP to the visual grounding. Based on the CLIP-based architecture, we further propose single-source and multi-source curriculum adapting algorithms, which can progressively find more reliable pseudo-labels to learn an optimal model, thereby achieving a balance between reliability and diversity for the pseudo-language labels. Our method outperforms the current state-of-the-art unsupervised method by a significant margin on RefCOCO/+/g datasets in both single-source and multi-source scenarios, with improvements ranging from 6.78% to 10.67% and 11.39% to 14.87%, respectively. Furthermore, our approach even outperforms existing weakly supervised methods. The code and models will be available at https://github.com/linhuixiao/CLIP-VG.
comment: 13 pages, 11 figures. Code will be released at https://github.com/linhuixiao/CLIP-VG
♻ ☆ Natural & Adversarial Bokeh Rendering via Circle-of-Confusion Predictive Network
Bokeh effect is a natural shallow depth-of-field phenomenon that blurs the out-of-focus part in photography. In recent years, a series of works have proposed automatic and realistic bokeh rendering methods for artistic and aesthetic purposes. They usually employ cutting-edge data-driven deep generative networks with complex training strategies and network architectures. However, these works neglect that the bokeh effect, as a real phenomenon, can inevitably affect the subsequent visual intelligent tasks like recognition, and their data-driven nature prevents them from studying the influence of bokeh-related physical parameters (i.e., depth-of-the-field) on the intelligent tasks. To fill this gap, we study a totally new problem, i.e., natural & adversarial bokeh rendering, which consists of two objectives: rendering realistic and natural bokeh and fooling the visual perception models (i.e., bokeh-based adversarial attack). To this end, beyond the pure data-driven solution, we propose a hybrid alternative by taking the respective advantages of data-driven and physical-aware methods. Specifically, we propose the circle-of-confusion predictive network (CoCNet) by taking the all-in-focus image and depth image as inputs to estimate circle-of-confusion parameters for each pixel, which are employed to render the final image through a well-known physical model of bokeh. With the hybrid solution, our method could achieve more realistic rendering results with the naive training strategy and a much lighter network. Moreover, we propose the adversarial bokeh attack by fixing the CoCNet while optimizing the depth map w.r.t the visual perception tasks. Then, we are able to study the vulnerability of deep neural networks according to the depth variations in the real world.
comment: 11 pages
♻ ☆ Literature Review: Computer Vision Applications in Transportation Logistics and Warehousing
Computer vision applications in transportation logistics and warehousing have a huge potential for process automation. We present a structured literature review on research in the field to help leverage this potential. The literature is categorized w.r.t. the application, i.e. the task it tackles and w.r.t. the computer vision techniques that are used. Regarding applications, we subdivide the literature in two areas: Monitoring, i.e. observing and retrieving relevant information from the environment, and manipulation, where approaches are used to analyze and interact with the environment. Additionally, we point out directions for future research and link to recent developments in computer vision that are suitable for application in logistics. Finally, we present an overview of existing datasets and industrial solutions. The results of our analysis are also available online at https://a-nau.github.io/cv-in-logistics.
♻ ☆ Reducing Predictive Feature Suppression in Resource-Constrained Contrastive Image-Caption Retrieval
To train image-caption retrieval (ICR) methods, contrastive loss functions are a common choice for optimization functions. Unfortunately, contrastive ICR methods are vulnerable to predictive feature suppression. Predictive features are features that correctly indicate the similarity between a query and a candidate item. However, in the presence of multiple predictive features during training, encoder models tend to suppress redundant predictive features, since these features are not needed to learn to discriminate between positive and negative pairs. While some predictive features are redundant during training, these features might be relevant during evaluation. We introduce an approach to reduce predictive feature suppression for resource-constrained ICR methods: latent target decoding (LTD). We add an additional decoder to the contrastive ICR framework, to reconstruct the input caption in a latent space of a general-purpose sentence encoder, which prevents the image and caption encoder from suppressing predictive features. We implement the LTD objective as an optimization constraint, to ensure that the reconstruction loss is below a bound value while primarily optimizing for the contrastive loss. Importantly, LTD does not depend on additional training data or expensive (hard) negative mining strategies. Our experiments show that, unlike reconstructing the input caption in the input space, LTD reduces predictive feature suppression, measured by obtaining higher recall@k, r-precision, and nDCG scores than a contrastive ICR baseline. Moreover, we show that LTD should be implemented as an optimization constraint instead of a dual optimization objective. Finally, we show that LTD can be used with different contrastive learning losses and a wide variety of resource-constrained ICR methods.
comment: Published in Transactions on Machine Learning Research OpenReview: https://openreview.net/forum?id=T1XtOqrVKn Code: https://github.com/MauritsBleeker/reducing-predictive-feature-suppression. Video: https://www.youtube.com/watch?v=oxa5AbGrKCY
♻ ☆ Digitization of Pathology Labs: A Review of Lessons Learned
Pathology laboratories are increasingly using digital workflows. This has the potential of increasing lab efficiency, but the digitization process also involves major challenges. Several reports have been published describing the individual experiences of specific laboratories with the digitization process. However, a comprehensive overview of the lessons learned is still lacking. We provide an overview of the lessons learned for different aspects of the digitization process, including digital case management, digital slide reading, and computer-aided slide reading. We also cover metrics used for monitoring performance and pitfalls and corresponding values observed in practice. The overview is intended to help pathologists, IT decision-makers, and administrators to benefit from the experiences of others and to implement the digitization process in an optimal way to make their own laboratory future-proof.
comment: 22 pages, 1 figure; corrected typo
♻ ☆ SSIVD-Net: A Novel Salient Super Image Classification & Detection Technique for Weaponized Violence
Detection of violence and weaponized violence in closed-circuit television (CCTV) footage requires a comprehensive approach. In this work, we introduce the \emph{Smart-City CCTV Violence Detection (SCVD)} dataset, specifically designed to facilitate the learning of weapon distribution in surveillance videos. To tackle the complexities of analyzing 3D surveillance video for violence recognition tasks, we propose a novel technique called, \emph{SSIVD-Net} (\textbf{S}alient-\textbf{S}uper-\textbf{I}mage for \textbf{V}iolence \textbf{D}etection). Our method reduces 3D video data complexity, dimensionality, and information loss while improving inference, performance, and explainability through the use of Salient-Super-Image representations. Considering the scalability and sustainability requirements of futuristic smart cities, the authors introduce the \emph{Salient-Classifier}, a novel architecture combining a kernelized approach with a residual learning strategy. We evaluate variations of SSIVD-Net and Salient Classifier on our SCVD dataset and benchmark against state-of-the-art (SOTA) models commonly employed in violence detection. Our approach exhibits significant improvements in detecting both weaponized and non-weaponized violence instances. By advancing the SOTA in violence detection, our work offers a practical and scalable solution suitable for real-world applications. The proposed methodology not only addresses the challenges of violence detection in CCTV footage but also contributes to the understanding of weapon distribution in smart surveillance. Ultimately, our research findings should enable smarter and more secure cities, as well as enhance public safety measures.
comment: 5 tables, 3 figures. Submitted to Machine Vision and Applications (Springer Nature) Journal
♻ ☆ Towards Accurate Data-free Quantization for Diffusion Models
In this paper, we propose an accurate data-free post-training quantization framework of diffusion models (ADP-DM) for efficient image generation. Conventional data-free quantization methods learn shared quantization functions for tensor discretization regardless of the generation timesteps, while the activation distribution differs significantly across various timesteps. The calibration images are acquired in random timesteps which fail to provide sufficient information for generalizable quantization function learning. Both issues cause sizable quantization errors with obvious image generation performance degradation. On the contrary, we design group-wise quantization functions for activation discretization in different timesteps and sample the optimal timestep for informative calibration image generation, so that our quantized diffusion model can reduce the discretization errors with negligible computational overhead. Specifically, we partition the timesteps according to the importance weights of quantization functions in different groups, which are optimized by differentiable search algorithms. We also select the optimal timestep for calibration image generation by structural risk minimizing principle in order to enhance the generalization ability in the deployment of quantized diffusion model. Extensive experimental results show that our method outperforms the state-of-the-art post-training quantization of diffusion model by a sizable margin with similar computational cost.
♻ ☆ Exploring Object-Centric Temporal Modeling for Efficient Multi-View 3D Object Detection
In this paper, we propose a long-sequence modeling framework, named StreamPETR, for multi-view 3D object detection. Built upon the sparse query design in the PETR series, we systematically develop an object-centric temporal mechanism. The model is performed in an online manner and the long-term historical information is propagated through object queries frame by frame. Besides, we introduce a motion-aware layer normalization to model the movement of the objects. StreamPETR achieves significant performance improvements only with negligible computation cost, compared to the single-frame baseline. On the standard nuScenes benchmark, it is the first online multi-view method that achieves comparable performance (67.6% NDS & 65.3% AMOTA) with lidar-based methods. The lightweight version realizes 45.0% mAP and 31.7 FPS, outperforming the state-of-the-art method (SOLOFusion) by 2.3% mAP and 1.8x faster FPS. Code has been available at https://github.com/exiawsh/StreamPETR.git.
♻ ☆ Occ-BEV: Multi-Camera Unified Pre-training via 3D Scene Reconstruction
Multi-camera 3D perception has emerged as a prominent research field in autonomous driving, offering a viable and cost-effective alternative to LiDAR-based solutions. However, existing multi-camera algorithms primarily rely on monocular image pre-training, which overlooks the spatial and temporal correlations among different camera views. To address this limitation, we propose the first multi-camera unified pre-training framework called Occ-BEV, which involves initially reconstructing the 3D scene as the foundational stage and subsequently fine-tuning the model on downstream tasks. Specifically, a 3D decoder is designed for leveraging Bird's Eye View (BEV) features from multi-view images to predict the 3D geometric occupancy to enable the model to capture a more comprehensive understanding of the 3D environment. A significant benefit of Occ-BEV is its capability of utilizing a considerable volume of unlabeled image-LiDAR pairs for pre-training purposes. The proposed multi-camera unified pre-training framework demonstrates promising results in key tasks such as multi-camera 3D object detection and surrounding semantic scene completion. When compared to monocular pre-training methods on the nuScenes dataset, Occ-BEV shows a significant improvement of about 2.0% in mAP and 2.0% in NDS for multi-camera 3D object detection, as well as a 3% increase in mIoU for surrounding semantic scene completion. Codes are publicly available at https://github.com/chaytonmin/Occ-BEV.
comment: 8 pages, 5 figures
♻ ☆ Cross-Modal Causal Relational Reasoning for Event-Level Visual Question Answering
Existing visual question answering methods often suffer from cross-modal spurious correlations and oversimplified event-level reasoning processes that fail to capture event temporality, causality, and dynamics spanning over the video. In this work, to address the task of event-level visual question answering, we propose a framework for cross-modal causal relational reasoning. In particular, a set of causal intervention operations is introduced to discover the underlying causal structures across visual and linguistic modalities. Our framework, named Cross-Modal Causal RelatIonal Reasoning (CMCIR), involves three modules: i) Causality-aware Visual-Linguistic Reasoning (CVLR) module for collaboratively disentangling the visual and linguistic spurious correlations via front-door and back-door causal interventions; ii) Spatial-Temporal Transformer (STT) module for capturing the fine-grained interactions between visual and linguistic semantics; iii) Visual-Linguistic Feature Fusion (VLFF) module for learning the global semantic-aware visual-linguistic representations adaptively. Extensive experiments on four event-level datasets demonstrate the superiority of our CMCIR in discovering visual-linguistic causal structures and achieving robust event-level visual question answering. The datasets, code, and models are available at https://github.com/HCPLab-SYSU/CMCIR.
comment: 17 pages, 9 figures. This work has been accepted by IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI). The datasets, code and models are available at https://github.com/HCPLab-SYSU/CMCIR
♻ ☆ PCT-CycleGAN: Paired Complementary Temporal Cycle-Consistent Adversarial Networks for Radar-Based Precipitation Nowcasting
The precipitation nowcasting methods have been elaborated over the centuries because rain has a crucial impact on human life. Not only quantitative precipitation forecast (QPF) models and convolutional long short-term memory (ConvLSTM), but also various sophisticated methods such as the latest MetNet-2 are emerging. In this paper, we propose a paired complementary temporal cycle-consistent adversarial networks (PCT-CycleGAN) for radar-based precipitation nowcasting, inspired by cycle-consistent adversarial networks (CycleGAN), which shows strong performance in image-to-image translation. PCT-CycleGAN generates temporal causality using two generator networks with forward and backward temporal dynamics in paired complementary cycles. Each generator network learns a huge number of one-to-one mappings about time-dependent radar-based precipitation data to approximate a mapping function representing the temporal dynamics in each direction. To create robust temporal causality between paired complementary cycles, novel connection loss is proposed. And torrential loss to cover exceptional heavy rain events is also proposed. The generator network learning forward temporal dynamics in PCT-CycleGAN generates radar-based precipitation data 10 minutes from the current time. Also, it provides a reliable prediction of up to 2 hours with iterative forecasting. The superiority of PCT-CycleGAN is demonstrated through qualitative and quantitative comparisons with several previous methods.
♻ ☆ CountingMOT: Joint Counting, Detection and Re-Identification for Multiple Object Tracking
The recent trend in multiple object tracking (MOT) is jointly solving detection and tracking, where object detection and appearance feature (or motion) are learned simultaneously. Despite competitive performance, in crowded scenes, joint detection and tracking usually fail to find accurate object associations due to missed or false detections. In this paper, we jointly model counting, detection and re-identification in an end-to-end framework, named CountingMOT, tailored for crowded scenes. By imposing mutual object-count constraints between detection and counting, the CountingMOT tries to find a balance between object detection and crowd density map estimation, which can help it to recover missed detections or reject false detections. Our approach is an attempt to bridge the gap of object detection, counting, and re-Identification. This is in contrast to prior MOT methods that either ignore the crowd density and thus are prone to failure in crowded scenes, or depend on local correlations to build a graphical relationship for matching targets. The proposed MOT tracker can perform online and real-time tracking, and achieves the state-of-the-art results on public benchmarks MOT16 (MOTA of 79.7), MOT17 (MOTA of 81.3%) and MOT20 (MOTA of 78.9%).
♻ ☆ ImplantFormer: Vision Transformer based Implant Position Regression Using Dental CBCT Data
Implant prosthesis is the most appropriate treatment for dentition defect or dentition loss, which usually involves a surgical guide design process to decide the implant position. However, such design heavily relies on the subjective experiences of dentists. In this paper, a transformer-based Implant Position Regression Network, ImplantFormer, is proposed to automatically predict the implant position based on the oral CBCT data. We creatively propose to predict the implant position using the 2D axial view of the tooth crown area and fit a centerline of the implant to obtain the actual implant position at the tooth root. Convolutional stem and decoder are designed to coarsely extract image features before the operation of patch embedding and integrate multi-level feature maps for robust prediction, respectively. As both long-range relationship and local features are involved, our approach can better represent global information and achieves better location performance. Extensive experiments on a dental implant dataset through five-fold cross-validation demonstrated that the proposed ImplantFormer achieves superior performance than existing methods.
♻ ☆ Revisit Weakly-Supervised Audio-Visual Video Parsing from the Language Perspective
We focus on the weakly-supervised audio-visual video parsing task (AVVP), which aims to identify and locate all the events in audio/visual modalities. Previous works only concentrate on video-level overall label denoising across modalities, but overlook the segment-level label noise, where adjacent video segments (i.e., 1-second video clips) may contain different events. However, recognizing events in the segment is challenging because its label could be any combination of events that occur in the video. To address this issue, we consider tackling AVVP from the language perspective, since language could freely describe how various events appear in each segment beyond fixed labels. Specifically, we design language prompts to describe all cases of event appearance for each video. Then, the similarity between language prompts and segments is calculated, where the event of the most similar prompt is regarded as the segment-level label. In addition, to deal with the mislabeled segments, we propose to perform dynamic re-weighting on the unreliable segments to adjust their labels. Experiments show that our simple yet effective approach outperforms state-of-the-art methods by a large margin.
♻ ☆ HICO-DET-SG and V-COCO-SG: New Data Splits for Evaluating the Systematic Generalization Performance of Human-Object Interaction Detection Models
Human-Object Interaction (HOI) detection is a task to localize humans and objects in an image and predict the interactions in human-object pairs. In real-world scenarios, HOI detection models are required systematic generalization, i.e., generalization to novel combinations of objects and interactions, because the train data are expected to cover a limited portion of all possible combinations. However, to our knowledge, no open benchmarks or previous work exist for evaluating the systematic generalization performance of HOI detection models. To address this issue, we created two new sets of HOI detection data splits named HICO-DET-SG and V-COCO-SG based on the HICO-DET and V-COCO datasets, respectively. When evaluated on the new data splits, the representative HOI detection models performed much more poorly than when evaluated on the original splits. This reveals that systematic generalization is a challenging goal in HOI detection. By analyzing the evaluation results, we also gain insights for improving the systematic generalization performance and identify four possible future research directions. We hope that our new data splits and presented analysis will encourage further research on systematic generalization in HOI detection.
comment: 19 pages, 3 figures, 4 tables
♻ ☆ Position-guided Text Prompt for Vision-Language Pre-training
Vision-Language Pre-Training (VLP) has shown promising capabilities to align image and text pairs, facilitating a broad variety of cross-modal learning tasks. However, we observe that VLP models often lack the visual grounding/localization capability which is critical for many downstream tasks such as visual reasoning. In this work, we propose a novel Position-guided Text Prompt (PTP) paradigm to enhance the visual grounding ability of cross-modal models trained with VLP. Specifically, in the VLP phase, PTP divides the image into $N\times N$ blocks, and identifies the objects in each block through the widely used object detector in VLP. It then reformulates the visual grounding task into a fill-in-the-blank problem given a PTP by encouraging the model to predict the objects in the given blocks or regress the blocks of a given object, e.g. filling `P" or ``O" in aPTP ``The block P has a O". This mechanism improves the visual grounding capability of VLP models and thus helps them better handle various downstream tasks. By introducing PTP into several state-of-the-art VLP frameworks, we observe consistently significant improvements across representative cross-modal learning model architectures and several benchmarks, e.g. zero-shot Flickr30K Retrieval (+4.8 in average recall@1) for ViLT \cite{vilt} baseline, and COCO Captioning (+5.3 in CIDEr) for SOTA BLIP \cite{blip} baseline. Moreover, PTP achieves comparable results with object-detector based methods, and much faster inference speed since PTP discards its object detector for inference while the later cannot. Our code and pre-trained weight will be released at \url{https://github.com/sail-sg/ptp}.
comment: Camera-ready version, code is in https://github.com/sail-sg/ptp
♻ ☆ Estimating 3D Dental Structures using Simulated Panoramic Radiographs and Neural Ray Tracing
Panoramic radiography (Panoramic X-ray, PX) is a widely used imaging modality for dental examination. Since PX only provides 2D flattened views of the oral structure, its applicability is limited as compared to 3D Cone-beam computed tomography (CBCT). In this paper, we propose a framework to estimate CBCT-like 3D structures from real-world PX. Our framework tackles full 3D reconstruction for varying subjects (patients) where each reconstruction is based only on a single panoramic image. We create an intermediate representation called simulated PX (SimPX) from CBCT data which is based both on the Beer-Lambert law of X-ray rendering and rotational principles of PX imaging. SimPX aims at not only truthfully simulating PX, but also facilitates the reverting process back to 3D data. We propose a novel neural model based on ray tracing which exploits both global and local input features to convert SimPX to 3D output. At inference, a real PX image is translated to a SimPX-style image with semantic regularization, and the translated image is processed by generation/refinement modules to produce high-quality outputs. Experiments show that our method outperforms prior state-of-the-art in reconstruction tasks both quantitatively and qualitatively. Our method does not require any prior information such as the shape of dental arches, nor the matched PX-CBCT dataset for training, which is difficult to obtain in clinical practice.
comment: 16 pages, 11 figures
♻ ☆ TextDiffuser: Diffusion Models as Text Painters
Diffusion models have gained increasing attention for their impressive generation abilities but currently struggle with rendering accurate and coherent text. To address this issue, we introduce TextDiffuser, focusing on generating images with visually appealing text that is coherent with backgrounds. TextDiffuser consists of two stages: first, a Transformer model generates the layout of keywords extracted from text prompts, and then diffusion models generate images conditioned on the text prompt and the generated layout. Additionally, we contribute the first large-scale text images dataset with OCR annotations, MARIO-10M, containing 10 million image-text pairs with text recognition, detection, and character-level segmentation annotations. We further collect the MARIO-Eval benchmark to serve as a comprehensive tool for evaluating text rendering quality. Through experiments and user studies, we show that TextDiffuser is flexible and controllable to create high-quality text images using text prompts alone or together with text template images, and conduct text inpainting to reconstruct incomplete images with text. The code, model, and dataset will be available at \url{https://aka.ms/textdiffuser}.
♻ ☆ A2B: Anchor to Barycentric Coordinate for Robust Correspondence
There is a long-standing problem of repeated patterns in correspondence problems, where mismatches frequently occur because of inherent ambiguity. The unique position information associated with repeated patterns makes coordinate representations a useful supplement to appearance representations for improving feature correspondences. However, the issue of appropriate coordinate representation has remained unresolved. In this study, we demonstrate that geometric-invariant coordinate representations, such as barycentric coordinates, can significantly reduce mismatches between features. The first step is to establish a theoretical foundation for geometrically invariant coordinates. We present a seed matching and filtering network (SMFNet) that combines feature matching and consistency filtering with a coarse-to-fine matching strategy in order to acquire reliable sparse correspondences. We then introduce DEGREE, a novel anchor-to-barycentric (A2B) coordinate encoding approach, which generates multiple affine-invariant correspondence coordinates from paired images. DEGREE can be used as a plug-in with standard descriptors, feature matchers, and consistency filters to improve the matching quality. Extensive experiments in synthesized indoor and outdoor datasets demonstrate that DEGREE alleviates the problem of repeated patterns and helps achieve state-of-the-art performance. Furthermore, DEGREE also reports competitive performance in the third Image Matching Challenge at CVPR 2021. This approach offers a new perspective to alleviate the problem of repeated patterns and emphasizes the importance of choosing coordinate representations for feature correspondences.
comment: Accepted by International Journal of Computer Vision
♻ ☆ Incremental Learning of Structured Memory via Closed-Loop Transcription
This work proposes a minimal computational model for learning structured memories of multiple object classes in an incremental setting. Our approach is based on establishing a closed-loop transcription between the classes and a corresponding set of subspaces, known as a linear discriminative representation, in a low-dimensional feature space. Our method is simpler than existing approaches for incremental learning, and more efficient in terms of model size, storage, and computation: it requires only a single, fixed-capacity autoencoding network with a feature space that is used for both discriminative and generative purposes. Network parameters are optimized simultaneously without architectural manipulations, by solving a constrained minimax game between the encoding and decoding maps over a single rate reduction-based objective. Experimental results show that our method can effectively alleviate catastrophic forgetting, achieving significantly better performance than prior work of generative replay on MNIST, CIFAR-10, and ImageNet-50, despite requiring fewer resources. Source code can be found at https://github.com/tsb0601/i-CTRL
comment: 20 pages
♻ ☆ PELE scores: Pelvic X-ray Landmark Detection by Pelvis Extraction and Enhancement
The pelvis, the lower part of the trunk, supports and balances the trunk. Landmark detection from a pelvic X-ray (PXR) facilitates downstream analysis and computer-assisted diagnosis and treatment of pelvic diseases. Although PXRs have the advantages of low radiation and reduced cost compared to computed tomography (CT) images, their 2D pelvis-tissue superposition of 3D structures confuses clinical decision-making. In this paper, we propose a PELvis Extraction (PELE) module that utilizes 3D prior anatomical knowledge in CT to guide and well isolate the pelvis from PXRs, thereby eliminating the influence of soft tissue. We conduct an extensive evaluation based on two public datasets and one private dataset, totaling 850 PXRs. The experimental results show that the proposed PELE module significantly improves the accuracy of PXRs landmark detection and achieves state-of-the-art performances in several benchmark metrics, thus better serving downstream tasks.
comment: will revise it and resubmit it again later
♻ ☆ Recognize Anything: A Strong Image Tagging Model
We present the Recognize Anything Model (RAM): a strong foundation model for image tagging. RAM can recognize any common category with high accuracy. RAM introduces a new paradigm for image tagging, leveraging large-scale image-text pairs for training instead of manual annotations. The development of RAM comprises four key steps. Firstly, annotation-free image tags are obtained at scale through automatic text semantic parsing. Subsequently, a preliminary model is trained for automatic annotation by unifying the caption and tagging tasks, supervised by the original texts and parsed tags, respectively. Thirdly, a data engine is employed to generate additional annotations and clean incorrect ones. Lastly, the model is retrained with the processed data and fine-tuned using a smaller but higher-quality dataset. We evaluate the tagging capabilities of RAM on numerous benchmarks and observe impressive zero-shot performance, significantly outperforming CLIP and BLIP. Remarkably, RAM even surpasses the fully supervised manners and exhibits competitive performance with the Google API. We are releasing the RAM at \url{https://recognize-anything.github.io/} to foster the advancements of large models in computer vision.
♻ ☆ IconShop: Text-Guided Vector Icon Synthesis with Autoregressive Transformers
Scalable Vector Graphics (SVG) is a popular vector image format that offers good support for interactivity and animation. Despite its appealing characteristics, creating custom SVG content can be challenging for users due to the steep learning curve required to understand SVG grammars or get familiar with professional editing software. Recent advancements in text-to-image generation have inspired researchers to explore vector graphics synthesis using either image-based methods (i.e., text -> raster image -> vector graphics) combining text-to-image generation models with image vectorization, or language-based methods (i.e., text -> vector graphics script) through pretrained large language models. However, these methods still suffer from limitations in terms of generation quality, diversity, and flexibility. In this paper, we introduce IconShop, a text-guided vector icon synthesis method using autoregressive transformers. The key to success of our approach is to sequentialize and tokenize SVG paths (and textual descriptions as guidance) into a uniquely decodable token sequence. With that, we are able to fully exploit the sequence learning power of autoregressive transformers, while enabling both unconditional and text-conditioned icon synthesis. Through standard training to predict the next token on a large-scale vector icon dataset accompanied by textural descriptions, the proposed IconShop consistently exhibits better icon synthesis capability than existing image-based and language-based methods both quantitatively and qualitatively. Meanwhile, we observe a dramatic improvement in generation diversity, which is validated by the objective Uniqueness and Novelty measures. More importantly, we demonstrate the flexibility of IconShop with multiple novel icon synthesis tasks, including icon editing, icon interpolation, icon semantic combination, and icon design auto-suggestion.
comment: Project Page: https://icon-shop.github.io/
♻ ☆ GLeaD: Improving GANs with A Generator-Leading Task CVPR2023
Generative adversarial network (GAN) is formulated as a two-player game between a generator (G) and a discriminator (D), where D is asked to differentiate whether an image comes from real data or is produced by G. Under such a formulation, D plays as the rule maker and hence tends to dominate the competition. Towards a fairer game in GANs, we propose a new paradigm for adversarial training, which makes G assign a task to D as well. Specifically, given an image, we expect D to extract representative features that can be adequately decoded by G to reconstruct the input. That way, instead of learning freely, D is urged to align with the view of G for domain classification. Experimental results on various datasets demonstrate the substantial superiority of our approach over the baselines. For instance, we improve the FID of StyleGAN2 from 4.30 to 2.55 on LSUN Bedroom and from 4.04 to 2.82 on LSUN Church. We believe that the pioneering attempt present in this work could inspire the community with better designed generator-leading tasks for GAN improvement.
comment: CVPR2023. Project page: https://ezioby.github.io/glead/ Code: https://github.com/EzioBy/glead/
♻ ☆ Vision-Centric BEV Perception: A Survey
In recent years, vision-centric Bird's Eye View (BEV) perception has garnered significant interest from both industry and academia due to its inherent advantages, such as providing an intuitive representation of the world and being conducive to data fusion. The rapid advancements in deep learning have led to the proposal of numerous methods for addressing vision-centric BEV perception challenges. However, there has been no recent survey encompassing this novel and burgeoning research field. To catalyze future research, this paper presents a comprehensive survey of the latest developments in vision-centric BEV perception and its extensions. It compiles and organizes up-to-date knowledge, offering a systematic review and summary of prevalent algorithms. Additionally, the paper provides in-depth analyses and comparative results on various BEV perception tasks, facilitating the evaluation of future works and sparking new research directions. Furthermore, the paper discusses and shares valuable empirical implementation details to aid in the advancement of related algorithms.
comment: project page at https://github.com/4DVLab/Vision-Centric-BEV-Perception; 22 pages, 15 figures
♻ ☆ Adversarial Infrared Blocks: A Black-box Attack to Thermal Infrared Detectors at Multiple Angles in Physical World
Infrared imaging systems have a vast array of potential applications in pedestrian detection and autonomous driving, and their safety performance is of great concern. However, few studies have explored the safety of infrared imaging systems in real-world settings. Previous research has used physical perturbations such as small bulbs and thermal "QR codes" to attack infrared imaging detectors, but such methods are highly visible and lack stealthiness. Other researchers have used hot and cold blocks to deceive infrared imaging detectors, but this method is limited in its ability to execute attacks from various angles. To address these shortcomings, we propose a novel physical attack called adversarial infrared blocks (AdvIB). By optimizing the physical parameters of the adversarial infrared blocks, this method can execute a stealthy black-box attack on thermal imaging system from various angles. We evaluate the proposed method based on its effectiveness, stealthiness, and robustness. Our physical tests show that the proposed method achieves a success rate of over 80% under most distance and angle conditions, validating its effectiveness. For stealthiness, our method involves attaching the adversarial infrared block to the inside of clothing, enhancing its stealthiness. Additionally, we test the proposed method on advanced detectors, and experimental results demonstrate an average attack success rate of 51.2%, proving its robustness. Overall, our proposed AdvIB method offers a promising avenue for conducting stealthy, effective and robust black-box attacks on thermal imaging system, with potential implications for real-world safety and security applications.
♻ ☆ PTQD: Accurate Post-Training Quantization for Diffusion Models
Diffusion models have recently dominated image synthesis and other related generative tasks. However, the iterative denoising process is expensive in computations at inference time, making diffusion models less practical for low-latency and scalable real-world applications. Post-training quantization of diffusion models can significantly reduce the model size and accelerate the sampling process without requiring any re-training. Nonetheless, applying existing post-training quantization methods directly to low-bit diffusion models can significantly impair the quality of generated samples. Specifically, for each denoising step, quantization noise leads to deviations in the estimated mean and mismatches with the predetermined variance schedule. Moreover, as the sampling process proceeds, the quantization noise may accumulate, resulting in a low signal-to-noise ratio (SNR) in late denoising steps. To address these challenges, we propose a unified formulation for the quantization noise and diffusion perturbed noise in the quantized denoising process. We first disentangle the quantization noise into its correlated and residual uncorrelated parts regarding its full-precision counterpart. The correlated part can be easily corrected by estimating the correlation coefficient. For the uncorrelated part, we calibrate the denoising variance schedule to absorb the excess variance resulting from quantization. Moreover, we propose a mixed-precision scheme to choose the optimal bitwidth for each denoising step, which prefers low bits to accelerate the early denoising steps while high bits maintain the high SNR for the late steps. Extensive experiments demonstrate that our method outperforms previous post-training quantized diffusion models in generating high-quality samples, with only a 0.06 increase in FID score compared to full-precision LDM-4 on ImageNet 256x256, while saving 19.9x bit operations.
comment: 18 pages, 14 figures
♻ ☆ HIPA: Hierarchical Patch Transformer for Single Image Super Resolution
Transformer-based architectures start to emerge in single image super resolution (SISR) and have achieved promising performance. Most existing Vision Transformers divide images into the same number of patches with a fixed size, which may not be optimal for restoring patches with different levels of texture richness. This paper presents HIPA, a novel Transformer architecture that progressively recovers the high resolution image using a hierarchical patch partition. Specifically, we build a cascaded model that processes an input image in multiple stages, where we start with tokens with small patch sizes and gradually merge to the full resolution. Such a hierarchical patch mechanism not only explicitly enables feature aggregation at multiple resolutions but also adaptively learns patch-aware features for different image regions, e.g., using a smaller patch for areas with fine details and a larger patch for textureless regions. Meanwhile, a new attention-based position encoding scheme for Transformer is proposed to let the network focus on which tokens should be paid more attention by assigning different weights to different tokens, which is the first time to our best knowledge. Furthermore, we also propose a new multi-reception field attention module to enlarge the convolution reception field from different branches. The experimental results on several public datasets demonstrate the superior performance of the proposed HIPA over previous methods quantitatively and qualitatively.
♻ ☆ Medical Visual Question Answering: A Survey
Medical Visual Question Answering~(VQA) is a combination of medical artificial intelligence and popular VQA challenges. Given a medical image and a clinically relevant question in natural language, the medical VQA system is expected to predict a plausible and convincing answer. Although the general-domain VQA has been extensively studied, the medical VQA still needs specific investigation and exploration due to its task features. In the first part of this survey, we collect and discuss the publicly available medical VQA datasets up-to-date about the data source, data quantity, and task feature. In the second part, we review the approaches used in medical VQA tasks. We summarize and discuss their techniques, innovations, and potential improvements. In the last part, we analyze some medical-specific challenges for the field and discuss future research directions. Our goal is to provide comprehensive and helpful information for researchers interested in the medical visual question answering field and encourage them to conduct further research in this field.
Information Retrieval 15
☆ MarineVRS: Marine Video Retrieval System with Explainability via Semantic Understanding
Building a video retrieval system that is robust and reliable, especially for the marine environment, is a challenging task due to several factors such as dealing with massive amounts of dense and repetitive data, occlusion, blurriness, low lighting conditions, and abstract queries. To address these challenges, we present MarineVRS, a novel and flexible video retrieval system designed explicitly for the marine domain. MarineVRS integrates state-of-the-art methods for visual and linguistic object representation to enable efficient and accurate search and analysis of vast volumes of underwater video data. In addition, unlike the conventional video retrieval system, which only permits users to index a collection of images or videos and search using a free-form natural language sentence, our retrieval system includes an additional Explainability module that outputs the segmentation masks of the objects that the input query referred to. This feature allows users to identify and isolate specific objects in the video footage, leading to more detailed analysis and understanding of their behavior and movements. Finally, with its adaptability, explainability, accuracy, and scalability, MarineVRS is a powerful tool for marine researchers and scientists to efficiently and accurately process vast amounts of data and gain deeper insights into the behavior and movements of marine species.
comment: Accepted to OCEANS 2023 Limerick. Website: https://marinevrs.hkustvgd.com/
☆ Constraint-based recommender system for crisis management simulations
In the context of the evacuation of populations, some citizens/volunteers may want and be able to participate in the evacuation of populations in difficulty by coming to lend a hand to emergency/evacuation vehicles with their own vehicles. One way of framing these impulses of solidarity would be to be able to list in real-time the citizens/volunteers available with their vehicles (land, sea, air, etc.), to be able to geolocate them according to the risk areas to be evacuated, and adding them to the evacuation/rescue vehicles. Because it is difficult to propose an effective real-time operational system on the field in a real crisis situation, in this work, we propose to add a module for recommending driver/vehicle pairs (with their specificities) to a system of crisis management simulation. To do that, we chose to model and develop an ontology-supported constraint-based recommender system for crisis management simulations.
☆ Embracing Uncertainty: Adaptive Vague Preference Policy Learning for Multi-round Conversational Recommendation
Conversational recommendation systems (CRS) effectively address information asymmetry by dynamically eliciting user preferences through multi-turn interactions. Existing CRS widely assumes that users have clear preferences. Under this assumption, the agent will completely trust the user feedback and treat the accepted or rejected signals as strong indicators to filter items and reduce the candidate space, which may lead to the problem of over-filtering. However, in reality, users' preferences are often vague and volatile, with uncertainty about their desires and changing decisions during interactions. To address this issue, we introduce a novel scenario called Vague Preference Multi-round Conversational Recommendation (VPMCR), which considers users' vague and volatile preferences in CRS.VPMCR employs a soft estimation mechanism to assign a non-zero confidence score for all candidate items to be displayed, naturally avoiding the over-filtering problem. In the VPMCR setting, we introduce an solution called Adaptive Vague Preference Policy Learning (AVPPL), which consists of two main components: Uncertainty-aware Soft Estimation (USE) and Uncertainty-aware Policy Learning (UPL). USE estimates the uncertainty of users' vague feedback and captures their dynamic preferences using a choice-based preferences extraction module and a time-aware decaying strategy. UPL leverages the preference distribution estimated by USE to guide the conversation and adapt to changes in users' preferences to make recommendations or ask for attributes. Our extensive experiments demonstrate the effectiveness of our method in the VPMCR scenario, highlighting its potential for practical applications and improving the overall performance and applicability of CRS in real-world settings, particularly for users with vague or dynamic preferences.
☆ RD-Suite: A Benchmark for Ranking Distillation
The distillation of ranking models has become an important topic in both academia and industry. In recent years, several advanced methods have been proposed to tackle this problem, often leveraging ranking information from teacher rankers that is absent in traditional classification settings. To date, there is no well-established consensus on how to evaluate this class of models. Moreover, inconsistent benchmarking on a wide range of tasks and datasets make it difficult to assess or invigorate advances in this field. This paper first examines representative prior arts on ranking distillation, and raises three questions to be answered around methodology and reproducibility. To that end, we propose a systematic and unified benchmark, Ranking Distillation Suite (RD-Suite), which is a suite of tasks with 4 large real-world datasets, encompassing two major modalities (textual and numeric) and two applications (standard distillation and distillation transfer). RD-Suite consists of benchmark results that challenge some of the common wisdom in the field, and the release of datasets with teacher scores and evaluation scripts for future research. RD-Suite paves the way towards better understanding of ranking distillation, facilities more research in this direction, and presents new challenges.
comment: 12 pages, 1 figure. arXiv admin note: text overlap with arXiv:2011.04006 by other authors
☆ Modeling Dual Period-Varying Preferences for Takeaway Recommendation
Takeaway recommender systems, which aim to accurately provide stores that offer foods meeting users' interests, have served billions of users in our daily life. Different from traditional recommendation, takeaway recommendation faces two main challenges: (1) Dual Interaction-Aware Preference Modeling. Traditional recommendation commonly focuses on users' single preferences for items while takeaway recommendation needs to comprehensively consider users' dual preferences for stores and foods. (2) Period-Varying Preference Modeling. Conventional recommendation generally models continuous changes in users' preferences from a session-level or day-level perspective. However, in practical takeaway systems, users' preferences vary significantly during the morning, noon, night, and late night periods of the day. To address these challenges, we propose a Dual Period-Varying Preference modeling (DPVP) for takeaway recommendation. Specifically, we design a dual interaction-aware module, aiming to capture users' dual preferences based on their interactions with stores and foods. Moreover, to model various preferences in different time periods of the day, we propose a time-based decomposition module as well as a time-aware gating mechanism. Extensive offline and online experiments demonstrate that our model outperforms state-of-the-art methods on real-world datasets and it is capable of modeling the dual period-varying preferences. Moreover, our model has been deployed online on Meituan Takeaway platform, leading to an average improvement in GMV (Gross Merchandise Value) of 0.70%.
☆ An Overview of Challenges in Egocentric Text-Video Retrieval CVPR 2023
Text-video retrieval contains various challenges, including biases coming from diverse sources. We highlight some of them supported by illustrations to open a discussion. Besides, we address one of the biases, frame length bias, with a simple method which brings a very incremental but promising increase. We conclude with future directions.
comment: 4 pages, CVPR 2023 Joint Ego4D&EPIC Workshop, Extended Abstract
☆ Phrase Retrieval for Open-Domain Conversational Question Answering with Conversational Dependency Modeling via Contrastive Learning ACL 2023
Open-Domain Conversational Question Answering (ODConvQA) aims at answering questions through a multi-turn conversation based on a retriever-reader pipeline, which retrieves passages and then predicts answers with them. However, such a pipeline approach not only makes the reader vulnerable to the errors propagated from the retriever, but also demands additional effort to develop both the retriever and the reader, which further makes it slower since they are not runnable in parallel. In this work, we propose a method to directly predict answers with a phrase retrieval scheme for a sequence of words, reducing the conventional two distinct subtasks into a single one. Also, for the first time, we study its capability for ODConvQA tasks. However, simply adopting it is largely problematic, due to the dependencies between previous and current turns in a conversation. To address this problem, we further introduce a novel contrastive learning strategy, making sure to reflect previous turns when retrieving the phrase for the current context, by maximizing representational similarities of consecutive turns in a conversation while minimizing irrelevant conversational contexts. We validate our model on two ODConvQA datasets, whose experimental results show that it substantially outperforms the relevant baselines with the retriever-reader. Code is available at: https://github.com/starsuzi/PRO-ConvQA.
comment: Findings of ACL 2023
☆ Set-to-Sequence Ranking-based Concept-aware Learning Path Recommendation
With the development of the online education system, personalized education recommendation has played an essential role. In this paper, we focus on developing path recommendation systems that aim to generating and recommending an entire learning path to the given user in each session. Noticing that existing approaches fail to consider the correlations of concepts in the path, we propose a novel framework named Set-to-Sequence Ranking-based Concept-aware Learning Path Recommendation (SRC), which formulates the recommendation task under a set-to-sequence paradigm. Specifically, we first design a concept-aware encoder module which can capture the correlations among the input learning concepts. The outputs are then fed into a decoder module that sequentially generates a path through an attention mechanism that handles correlations between the learning and target concepts. Our recommendation policy is optimized by policy gradient. In addition, we also introduce an auxiliary module based on knowledge tracing to enhance the model's stability by evaluating students' learning effects on learning concepts. We conduct extensive experiments on two real-world public datasets and one industrial dataset, and the experimental results demonstrate the superiority and effectiveness of SRC. Code will be available at https://gitee.com/mindspore/models/tree/master/research/recommend/SRC.
☆ SANGEET: A XML based Open Dataset for Research in Hindustani Sangeet
It is very important to access a rich music dataset that is useful in a wide variety of applications. Currently, available datasets are mostly focused on storing vocal or instrumental recording data and ignoring the requirement of its visual representation and retrieval. This paper attempts to build an XML-based public dataset, called SANGEET, that stores comprehensive information of Hindustani Sangeet (North Indian Classical Music) compositions written by famous musicologist Pt. Vishnu Narayan Bhatkhande. SANGEET preserves all the required information of any given composition including metadata, structural, notational, rhythmic, and melodic information in a standardized way for easy and efficient storage and extraction of musical information. The dataset is intended to provide the ground truth information for music information research tasks, thereby supporting several data-driven analysis from a machine learning perspective. We present the usefulness of the dataset by demonstrating its application on music information retrieval using XQuery, visualization through Omenad rendering system. Finally, we propose approaches to transform the dataset for performing statistical and machine learning tasks for a better understanding of Hindustani Sangeet. The dataset can be found at https://github.com/cmisra/Sangeet.
☆ Answering Compositional Queries with Set-Theoretic Embeddings
The need to compactly and robustly represent item-attribute relations arises in many important tasks, such as faceted browsing and recommendation systems. A popular machine learning approach for this task denotes that an item has an attribute by a high dot-product between vectors for the item and attribute -- a representation that is not only dense, but also tends to correct noisy and incomplete data. While this method works well for queries retrieving items by a single attribute (such as \emph{movies that are comedies}), we find that vector embeddings do not so accurately support compositional queries (such as movies that are comedies and British but not romances). To address these set-theoretic compositions, this paper proposes to replace vectors with box embeddings, a region-based representation that can be thought of as learnable Venn diagrams. We introduce a new benchmark dataset for compositional queries, and present experiments and analysis providing insights into the behavior of both. We find that, while vector and box embeddings are equally suited to single attribute queries, for compositional queries box embeddings provide substantial advantages over vectors, particularly at the moderate and larger retrieval set sizes that are most useful for users' search and browsing.
☆ PANE-GNN: Unifying Positive and Negative Edges in Graph Neural Networks for Recommendation
Recommender systems play a crucial role in addressing the issue of information overload by delivering personalized recommendations to users. In recent years, there has been a growing interest in leveraging graph neural networks (GNNs) for recommender systems, capitalizing on advancements in graph representation learning. These GNN-based models primarily focus on analyzing users' positive feedback while overlooking the valuable insights provided by their negative feedback. In this paper, we propose PANE-GNN, an innovative recommendation model that unifies Positive And Negative Edges in Graph Neural Networks for recommendation. By incorporating user preferences and dispreferences, our approach enhances the capability of recommender systems to offer personalized suggestions. PANE-GNN first partitions the raw rating graph into two distinct bipartite graphs based on positive and negative feedback. Subsequently, we employ two separate embeddings, the interest embedding and the disinterest embedding, to capture users' likes and dislikes, respectively. To facilitate effective information propagation, we design distinct message-passing mechanisms for positive and negative feedback. Furthermore, we introduce a distortion to the negative graph, which exclusively consists of negative feedback edges, for contrastive training. This distortion plays a crucial role in effectively denoising the negative feedback. The experimental results provide compelling evidence that PANE-GNN surpasses the existing state-of-the-art benchmark methods across four real-world datasets. These datasets include three commonly used recommender system datasets and one open-source short video recommendation dataset.
♻ ☆ PALR: Personalization Aware LLMs for Recommendation
Large language models (LLMs) have recently received significant attention for their exceptional capabilities. Despite extensive efforts in developing general-purpose LLMs that can be utilized in various natural language processing (NLP) tasks, there has been less research exploring their potential in recommender systems. In this paper, we propose a novel framework, named PALR, which aiming to combine user history behaviors (such as clicks, purchases, ratings, etc.) with LLMs to generate user preferred items. Specifically, we first use user/item interactions as guidance for candidate retrieval. Then we adopt a LLM-based ranking model to generate recommended items. Unlike existing approaches that typically adopt general-purpose LLMs for zero/few-shot recommendation testing or training on small-sized language models (with less than 1 billion parameters), which cannot fully elicit LLMs' reasoning abilities and leverage rich item side parametric knowledge, we fine-tune a 7 billion parameters LLM for the ranking purpose. This model takes retrieval candidates in natural language format as input, with instruction which explicitly asking to select results from input candidates during inference. Our experimental results demonstrate that our solution outperforms state-of-the-art models on various sequential recommendation tasks.
♻ ☆ Learning to Suggest Breaks: Sustainable Optimization of Long-Term User Engagement ICML 2023
Optimizing user engagement is a key goal for modern recommendation systems, but blindly pushing users towards increased consumption risks burn-out, churn, or even addictive habits. To promote digital well-being, most platforms now offer a service that periodically prompts users to take breaks. These, however, must be set up manually, and so may be suboptimal for both users and the system. In this paper, we study the role of breaks in recommendation, and propose a framework for learning optimal breaking policies that promote and sustain long-term engagement. Based on the notion that recommendation dynamics are susceptible to both positive and negative feedback, we cast recommendation as a Lotka-Volterra dynamical system, where breaking reduces to a problem of optimal control. We then give an efficient learning algorithm, provide theoretical guarantees, and empirically demonstrate the utility of our approach on semi-synthetic data.
comment: Accepted for publication in ICML 2023
♻ ☆ bgGLUE: A Bulgarian General Language Understanding Evaluation Benchmark ACL 2023
We present bgGLUE(Bulgarian General Language Understanding Evaluation), a benchmark for evaluating language models on Natural Language Understanding (NLU) tasks in Bulgarian. Our benchmark includes NLU tasks targeting a variety of NLP problems (e.g., natural language inference, fact-checking, named entity recognition, sentiment analysis, question answering, etc.) and machine learning tasks (sequence labeling, document-level classification, and regression). We run the first systematic evaluation of pre-trained language models for Bulgarian, comparing and contrasting results across the nine tasks in the benchmark. The evaluation results show strong performance on sequence labeling tasks, but there is a lot of room for improvement for tasks that require more complex reasoning. We make bgGLUE publicly available together with the fine-tuning and the evaluation code, as well as a public leaderboard at https://bgglue.github.io/, and we hope that it will enable further advancements in developing NLU models for Bulgarian.
comment: Accepted to ACL 2023 (Main Conference)
♻ ☆ DataFinder: Scientific Dataset Recommendation from Natural Language Descriptions ACL 2023
Modern machine learning relies on datasets to develop and validate research ideas. Given the growth of publicly available data, finding the right dataset to use is increasingly difficult. Any research question imposes explicit and implicit constraints on how well a given dataset will enable researchers to answer this question, such as dataset size, modality, and domain. We operationalize the task of recommending datasets given a short natural language description of a research idea, to help people find relevant datasets for their needs. Dataset recommendation poses unique challenges as an information retrieval problem; datasets are hard to directly index for search and there are no corpora readily available for this task. To facilitate this task, we build the DataFinder Dataset which consists of a larger automatically-constructed training set (17.5K queries) and a smaller expert-annotated evaluation set (392 queries). Using this data, we compare various information retrieval algorithms on our test set and present a superior bi-encoder retriever for text-based dataset recommendation. This system, trained on the DataFinder Dataset, finds more relevant search results than existing third-party dataset search engines. To encourage progress on dataset recommendation, we release our dataset and models to the public.
comment: To appear at ACL 2023. Code published at https://github.com/viswavi/datafinder
Machine Learning 150
☆ ModuleFormer: Learning Modular Large Language Models From Uncurated Data
Large Language Models (LLMs) have achieved remarkable results. But existing models are expensive to train and deploy, and it is also difficult to expand their knowledge beyond pre-training data without forgetting previous knowledge. This paper proposes a new neural network architecture, ModuleFormer, that leverages modularity to improve the efficiency and flexibility of large language models. ModuleFormer is based on the Sparse Mixture of Experts (SMoE). Unlike the previous SMoE-based modular language model [Gururangan et al., 2021], which requires domain-labeled data to learn domain-specific experts, ModuleFormer can induce modularity from uncurated data with its new load balancing and load concentration losses. ModuleFormer is a modular architecture that includes two different types of modules, new stick-breaking attention heads, and feedforward experts. Different modules are sparsely activated conditions on the input token during training and inference. In our experiment, we found that the modular architecture enables three important abilities for large pre-trained language models: 1) Efficiency, since ModuleFormer only activates a subset of its modules for each input token, thus it could achieve the same performance as dense LLMs with more than two times throughput; 2) Extendability, ModuleFormer is more immune to catastrophic forgetting than dense LLMs and can be easily extended with new modules to learn new knowledge that is not included in the training data; 3) Specialisation, finetuning ModuleFormer could specialize a subset of modules to the finetuning task, and the task-unrelated modules could be easily pruned for a lightweight deployment.
Transformers as Statisticians: Provable In-Context Learning with In-Context Algorithm Selection
Neural sequence models based on the transformer architecture have demonstrated remarkable \emph{in-context learning} (ICL) abilities, where they can perform new tasks when prompted with training and test examples, without any parameter update to the model. This work first provides a comprehensive statistical theory for transformers to perform ICL. Concretely, we show that transformers can implement a broad class of standard machine learning algorithms in context, such as least squares, ridge regression, Lasso, learning generalized linear models, and gradient descent on two-layer neural networks, with near-optimal predictive power on various in-context data distributions. Using an efficient implementation of in-context gradient descent as the underlying mechanism, our transformer constructions admit mild size bounds, and can be learned with polynomially many pretraining sequences. Building on these ``base'' ICL algorithms, intriguingly, we show that transformers can implement more complex ICL procedures involving \emph{in-context algorithm selection}, akin to what a statistician can do in real life -- A \emph{single} transformer can adaptively select different base ICL algorithms -- or even perform qualitatively different tasks -- on different input sequences, without any explicit prompting of the right algorithm or task. We both establish this in theory by explicit constructions, and also observe this phenomenon experimentally. In theory, we construct two general mechanisms for algorithm selection with concrete examples: pre-ICL testing, and post-ICL validation. As an example, we use the post-ICL validation mechanism to construct a transformer that can perform nearly Bayes-optimal ICL on a challenging task -- noisy linear models with mixed noise levels. Experimentally, we demonstrate the strong in-context algorithm selection capabilities of standard transformer architectures.
☆ GP-UNIT: Generative Prior for Versatile Unsupervised Image-to-Image Translation
Recent advances in deep learning have witnessed many successful unsupervised image-to-image translation models that learn correspondences between two visual domains without paired data. However, it is still a great challenge to build robust mappings between various domains especially for those with drastic visual discrepancies. In this paper, we introduce a novel versatile framework, Generative Prior-guided UNsupervised Image-to-image Translation (GP-UNIT), that improves the quality, applicability and controllability of the existing translation models. The key idea of GP-UNIT is to distill the generative prior from pre-trained class-conditional GANs to build coarse-level cross-domain correspondences, and to apply the learned prior to adversarial translations to excavate fine-level correspondences. With the learned multi-level content correspondences, GP-UNIT is able to perform valid translations between both close domains and distant domains. For close domains, GP-UNIT can be conditioned on a parameter to determine the intensity of the content correspondences during translation, allowing users to balance between content and style consistency. For distant domains, semi-supervised learning is explored to guide GP-UNIT to discover accurate semantic correspondences that are hard to learn solely from the appearance. We validate the superiority of GP-UNIT over state-of-the-art translation models in robust, high-quality and diversified translations between various domains through extensive experiments.
comment: Accepted by IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI). Code: https://github.com/williamyang1991/GP-UNIT Project page: https://www.mmlab-ntu.com/project/gpunit/. arXiv admin note: substantial text overlap with arXiv:2204.03641
☆ On the Reliability of Watermarks for Large Language Models
Large language models (LLMs) are now deployed to everyday use and positioned to produce large quantities of text in the coming decade. Machine-generated text may displace human-written text on the internet and has the potential to be used for malicious purposes, such as spearphishing attacks and social media bots. Watermarking is a simple and effective strategy for mitigating such harms by enabling the detection and documentation of LLM-generated text. Yet, a crucial question remains: How reliable is watermarking in realistic settings in the wild? There, watermarked text might be mixed with other text sources, paraphrased by human writers or other language models, and used for applications in a broad number of domains, both social and technical. In this paper, we explore different detection schemes, quantify their power at detecting watermarks, and determine how much machine-generated text needs to be observed in each scenario to reliably detect the watermark. We especially highlight our human study, where we investigate the reliability of watermarking when faced with human paraphrasing. We compare watermark-based detection to other detection strategies, finding overall that watermarking is a reliable solution, especially because of its sample complexity - for all attacks we consider, the watermark evidence compounds the more examples are given, and the watermark is eventually detected.
comment: 14 pages in the main body. Code is available at https://github.com/jwkirchenbauer/lm-watermarking
☆ Contrastive Lift: 3D Object Instance Segmentation by Slow-Fast Contrastive Fusion
Instance segmentation in 3D is a challenging task due to the lack of large-scale annotated datasets. In this paper, we show that this task can be addressed effectively by leveraging instead 2D pre-trained models for instance segmentation. We propose a novel approach to lift 2D segments to 3D and fuse them by means of a neural field representation, which encourages multi-view consistency across frames. The core of our approach is a slow-fast clustering objective function, which is scalable and well-suited for scenes with a large number of objects. Unlike previous approaches, our method does not require an upper bound on the number of objects or object tracking across frames. To demonstrate the scalability of the slow-fast clustering, we create a new semi-realistic dataset called the Messy Rooms dataset, which features scenes with up to 500 objects per scene. Our approach outperforms the state-of-the-art on challenging scenes from the ScanNet, Hypersim, and Replica datasets, as well as on our newly created Messy Rooms dataset, demonstrating the effectiveness and scalability of our slow-fast clustering method.
☆ Yet Another Algorithm for Supervised Principal Component Analysis: Supervised Linear Centroid-Encoder
We propose a new supervised dimensionality reduction technique called Supervised Linear Centroid-Encoder (SLCE), a linear counterpart of the nonlinear Centroid-Encoder (CE) \citep{ghosh2022supervised}. SLCE works by mapping the samples of a class to its class centroid using a linear transformation. The transformation is a projection that reconstructs a point such that its distance from the corresponding class centroid, i.e., centroid-reconstruction loss, is minimized in the ambient space. We derive a closed-form solution using an eigendecomposition of a symmetric matrix. We did a detailed analysis and presented some crucial mathematical properties of the proposed approach. %We also provide an iterative solution approach based solving the optimization problem using a descent method. We establish a connection between the eigenvalues and the centroid-reconstruction loss. In contrast to Principal Component Analysis (PCA) which reconstructs a sample in the ambient space, the transformation of SLCE uses the instances of a class to rebuild the corresponding class centroid. Therefore the proposed method can be considered a form of supervised PCA. Experimental results show the performance advantage of SLCE over other supervised methods.
comment: A novel algorithm for supervised PCA. 22 pages (including 2 reference pages), 8 figures and mathematical analysis of the proposed algorithm. The article is under review now
☆ Align, Distill, and Augment Everything All at Once for Imbalanced Semi-Supervised Learning
Addressing the class imbalance in long-tailed semi-supervised learning (SSL) poses a few significant challenges stemming from differences between the marginal distributions of unlabeled data and the labeled data, as the former is often unknown and potentially distinct from the latter. The first challenge is to avoid biasing the pseudo-labels towards an incorrect distribution, such as that of the labeled data or a balanced distribution, during training. However, we still wish to ensure a balanced unlabeled distribution during inference, which is the second challenge. To address both of these challenges, we propose a three-faceted solution: a flexible distribution alignment that progressively aligns the classifier from a dynamically estimated unlabeled prior towards a balanced distribution, a soft consistency regularization that exploits underconfident pseudo-labels discarded by threshold-based methods, and a schema for expanding the unlabeled set with input data from the labeled partition. This last facet comes in as a response to the commonly-overlooked fact that disjoint partitions of labeled and unlabeled data prevent the benefits of strong data augmentation on the labeled set. Our overall framework requires no additional training cycles, so it will align, distill, and augment everything all at once (ADALLO). Our extensive evaluations of ADALLO on imbalanced SSL benchmark datasets, including CIFAR10-LT, CIFAR100-LT, and STL10-LT with varying degrees of class imbalance, amount of labeled data, and distribution mismatch, demonstrate significant improvements in the performance of imbalanced SSL under large distribution mismatch, as well as competitiveness with state-of-the-art methods when the labeled and unlabeled data follow the same marginal distribution. Our code will be released upon paper acceptance.
comment: Under review, 12 pages
☆ Goal-conditioned GFlowNets for Controllable Multi-Objective Molecular Design
In recent years, in-silico molecular design has received much attention from the machine learning community. When designing a new compound for pharmaceutical applications, there are usually multiple properties of such molecules that need to be optimised: binding energy to the target, synthesizability, toxicity, EC50, and so on. While previous approaches have employed a scalarization scheme to turn the multi-objective problem into a preference-conditioned single objective, it has been established that this kind of reduction may produce solutions that tend to slide towards the extreme points of the objective space when presented with a problem that exhibits a concave Pareto front. In this work we experiment with an alternative formulation of goal-conditioned molecular generation to obtain a more controllable conditional model that can uniformly explore solutions along the entire Pareto front.
comment: 14 pages
☆ Revisiting Out-of-distribution Robustness in NLP: Benchmark, Analysis, and LLMs Evaluations
This paper reexamines the research on out-of-distribution (OOD) robustness in the field of NLP. We find that the distribution shift settings in previous studies commonly lack adequate challenges, hindering the accurate evaluation of OOD robustness. To address these issues, we propose a benchmark construction protocol that ensures clear differentiation and challenging distribution shifts. Then we introduce BOSS, a Benchmark suite for Out-of-distribution robustneSS evaluation covering 5 tasks and 20 datasets. Based on BOSS, we conduct a series of experiments on pre-trained language models for analysis and evaluation of OOD robustness. First, for vanilla fine-tuning, we examine the relationship between in-distribution (ID) and OOD performance. We identify three typical types that unveil the inner learning mechanism, which could potentially facilitate the forecasting of OOD robustness, correlating with the advancements on ID datasets. Then, we evaluate 5 classic methods on BOSS and find that, despite exhibiting some effectiveness in specific cases, they do not offer significant improvement compared to vanilla fine-tuning. Further, we evaluate 5 LLMs with various adaptation paradigms and find that when sufficient ID data is available, fine-tuning domain-specific models outperform LLMs on ID examples significantly. However, in the case of OOD instances, prioritizing LLMs with in-context learning yields better results. We identify that both fine-tuned small models and LLMs face challenges in effectively addressing downstream tasks. The code is public at \url{https://github.com/lifan-yuan/OOD_NLP}.
comment: Code is available at \url{https://github.com/lifan-yuan/OOD_NLP}
☆ Uncovering solutions from data corrupted by systematic errors: A physics-constrained convolutional neural network approach
Information on natural phenomena and engineering systems is typically contained in data. Data can be corrupted by systematic errors in models and experiments. In this paper, we propose a tool to uncover the spatiotemporal solution of the underlying physical system by removing the systematic errors from data. The tool is the physics-constrained convolutional neural network (PC-CNN), which combines information from both the systems governing equations and data. We focus on fundamental phenomena that are modelled by partial differential equations, such as linear convection, Burgers equation, and two-dimensional turbulence. First, we formulate the problem, describe the physics-constrained convolutional neural network, and parameterise the systematic error. Second, we uncover the solutions from data corrupted by large multimodal systematic errors. Third, we perform a parametric study for different systematic errors. We show that the method is robust. Fourth, we analyse the physical properties of the uncovered solutions. We show that the solutions inferred from the PC-CNN are physical, in contrast to the data corrupted by systematic errors that does not fulfil the governing equations. This work opens opportunities for removing epistemic errors from models, and systematic errors from measurements.
☆ Language Models Get a Gender Makeover: Mitigating Gender Bias with Few-Shot Data Interventions ACL 2023
Societal biases present in pre-trained large language models are a critical issue as these models have been shown to propagate biases in countless downstream applications, rendering them unfair towards specific groups of people. Since large-scale retraining of these models from scratch is both time and compute-expensive, a variety of approaches have been previously proposed that de-bias a pre-trained model. While the majority of current state-of-the-art debiasing methods focus on changes to the training regime, in this paper, we propose data intervention strategies as a powerful yet simple technique to reduce gender bias in pre-trained models. Specifically, we empirically show that by fine-tuning a pre-trained model on only 10 de-biased (intervened) training examples, the tendency to favor any gender is significantly reduced. Since our proposed method only needs a few training examples, our few-shot debiasing approach is highly feasible and practical. Through extensive experimentation, we show that our debiasing technique performs better than competitive state-of-the-art baselines with minimal loss in language modeling ability.
comment: Accepted to ACL 2023 Main Conference
☆ Generalization Across Observation Shifts in Reinforcement Learning
Learning policies which are robust to changes in the environment are critical for real world deployment of Reinforcement Learning agents. They are also necessary for achieving good generalization across environment shifts. We focus on bisimulation metrics, which provide a powerful means for abstracting task relevant components of the observation and learning a succinct representation space for training the agent using reinforcement learning. In this work, we extend the bisimulation framework to also account for context dependent observation shifts. Specifically, we focus on the simulator based learning setting and use alternate observations to learn a representation space which is invariant to observation shifts using a novel bisimulation based objective. This allows us to deploy the agent to varying observation settings during test time and generalize to unseen scenarios. We further provide novel theoretical bounds for simulator fidelity and performance transfer guarantees for using a learnt policy to unseen shifts. Empirical analysis on the high-dimensional image based control domains demonstrates the efficacy of our method.
☆ Proximity-Informed Calibration for Deep Neural Networks
Confidence calibration is central to providing accurate and interpretable uncertainty estimates, especially under safety-critical scenarios. However, we find that existing calibration algorithms often overlook the issue of proximity bias, a phenomenon where models tend to be more overconfident in low proximity data (i.e., lying in the sparse region of the data distribution) compared to high proximity samples, and thus suffer from inconsistent miscalibration across different proximity samples. We examine the problem over pretrained ImageNet models and observe that: 1) Proximity bias exists across a wide variety of model architectures and sizes; 2) Transformer-based models are more susceptible to proximity bias than CNN-based models; 3) Proximity bias persists even after performing popular calibration algorithms like temperature scaling; 4) Models tend to overfit more heavily on low proximity samples than on high proximity samples. Motivated by the empirical findings, we propose ProCal, a plug-and-play algorithm with a theoretical guarantee to adjust sample confidence based on proximity. To further quantify the effectiveness of calibration algorithms in mitigating proximity bias, we introduce proximity-informed expected calibration error (PIECE) with theoretical analysis. We show that ProCal is effective in addressing proximity bias and improving calibration on balanced, long-tail, and distribution-shift settings under four metrics over various model architectures.
☆ Divide and Repair: Using Options to Improve Performance of Imitation Learning Against Adversarial Demonstrations
We consider the problem of learning to perform a task from demonstrations given by teachers or experts, when some of the experts' demonstrations might be adversarial and demonstrate an incorrect way to perform the task. We propose a novel technique that can identify parts of demonstrated trajectories that have not been significantly modified by the adversary and utilize them for learning, using temporally extended policies or options. We first define a trajectory divergence measure based on the spatial and temporal features of demonstrated trajectories to detect and discard parts of the trajectories that have been significantly modified by an adversarial expert, and, could degrade the learner's performance, if used for learning, We then use an options-based algorithm that partitions trajectories and learns only from the parts of trajectories that have been determined as admissible. We provide theoretical results of our technique to show that repairing partial trajectories improves the sample efficiency of the demonstrations without degrading the learner's performance. We then evaluate the proposed algorithm for learning to play an Atari-like, computer-based game called LunarLander in the presence of different types and degrees of adversarial attacks of demonstrated trajectories. Our experimental results show that our technique can identify adversarially modified parts of the demonstrated trajectories and successfully prevent the learning performance from degrading due to adversarial demonstrations.
comment: 33 pages, 4 figures, 3 tables
☆ Recent applications of machine learning, remote sensing, and iot approaches in yield prediction: a critical review
The integration of remote sensing and machine learning in agriculture is transforming the industry by providing insights and predictions through data analysis. This combination leads to improved yield prediction and water management, resulting in increased efficiency, better yields, and more sustainable agricultural practices. Achieving the United Nations' Sustainable Development Goals, especially "zero hunger," requires the investigation of crop yield and precipitation gaps, which can be accomplished through, the usage of artificial intelligence (AI), machine learning (ML), remote sensing (RS), and the internet of things (IoT). By integrating these technologies, a robust agricultural mobile or web application can be developed, providing farmers and decision-makers with valuable information and tools for improving crop management and increasing efficiency. Several studies have investigated these new technologies and their potential for diverse tasks such as crop monitoring, yield prediction, irrigation management, etc. Through a critical review, this paper reviews relevant articles that have used RS, ML, cloud computing, and IoT in crop yield prediction. It reviews the current state-of-the-art in this field by critically evaluating different machine-learning approaches proposed in the literature for crop yield prediction and water management. It provides insights into how these methods can improve decision-making in agricultural production systems. This work will serve as a compendium for those interested in yield prediction in terms of primary literature but, most importantly, what approaches can be used for real-time and robust prediction.
comment: 35 pages, 12 figures, 14 tables
☆ ChatGPT is fun, but it is not funny! Humor is still challenging Large Language Models
Humor is a central aspect of human communication that has not been solved for artificial agents so far. Large language models (LLMs) are increasingly able to capture implicit and contextual information. Especially, OpenAI's ChatGPT recently gained immense public attention. The GPT3-based model almost seems to communicate on a human level and can even tell jokes. Humor is an essential component of human communication. But is ChatGPT really funny? We put ChatGPT's sense of humor to the test. In a series of exploratory experiments around jokes, i.e., generation, explanation, and detection, we seek to understand ChatGPT's capability to grasp and reproduce human humor. Since the model itself is not accessible, we applied prompt-based experiments. Our empirical evidence indicates that jokes are not hard-coded but mostly also not newly generated by the model. Over 90% of 1008 generated jokes were the same 25 Jokes. The system accurately explains valid jokes but also comes up with fictional explanations for invalid jokes. Joke-typical characteristics can mislead ChatGPT in the classification of jokes. ChatGPT has not solved computational humor yet but it can be a big leap toward "funny" machines.
☆ StudentEval: A Benchmark of Student-Written Prompts for Large Language Models of Code
Code LLMs are being rapidly deployed and there is evidence that they can make professional programmers more productive. Current benchmarks for code generation measure whether models generate correct programs given an expert prompt. In this paper, we present a new benchmark containing multiple prompts per problem, written by a specific population of non-expert prompters: beginning programmers. StudentEval contains 1,749 prompts for 48 problems, written by 80 students who have only completed one semester of Python programming. Our students wrote these prompts while working interactively with a Code LLM, and we observed very mixed success rates. We use StudentEval to evaluate 5 Code LLMs and find that StudentEval is a better discriminator of model performance than existing benchmarks. We analyze the prompts and find significant variation in students' prompting techniques. We also find that nondeterministic LLM sampling could mislead students into thinking that their prompts are more (or less) effective than they actually are, which has implications for how to teach with Code LLMs.
☆ Multi-Task Training with In-Domain Language Models for Diagnostic Reasoning
Generative artificial intelligence (AI) is a promising direction for augmenting clinical diagnostic decision support and reducing diagnostic errors, a leading contributor to medical errors. To further the development of clinical AI systems, the Diagnostic Reasoning Benchmark (DR.BENCH) was introduced as a comprehensive generative AI framework, comprised of six tasks representing key components in clinical reasoning. We present a comparative analysis of in-domain versus out-of-domain language models as well as multi-task versus single task training with a focus on the problem summarization task in DR.BENCH (Gao et al., 2023). We demonstrate that a multi-task, clinically trained language model outperforms its general domain counterpart by a large margin, establishing a new state-of-the-art performance, with a ROUGE-L score of 28.55. This research underscores the value of domain-specific training for optimizing clinical diagnostic reasoning tasks.
comment: Accepted to 2023 ClinicalNLP Workshop
☆ Convergence of SARSA with linear function approximation: The random horizon case
The reinforcement learning algorithm SARSA combined with linear function approximation has been shown to converge for infinite horizon discounted Markov decision problems (MDPs). In this paper, we investigate the convergence of the algorithm for random horizon MDPs, which has not previously been shown. We show, similar to earlier results for infinite horizon discounted MDPs, that if the behaviour policy is $\varepsilon$-soft and Lipschitz continuous with respect to the weight vector of the linear function approximation, with small enough Lipschitz constant, then the algorithm will converge with probability one when considering a random horizon MDP.
☆ On the Design Fundamentals of Diffusion Models: A Survey
Diffusion models are generative models, which gradually add and remove noise to learn the underlying distribution of training data for data generation. The components of diffusion models have gained significant attention with many design choices proposed. Existing reviews have primarily focused on higher-level solutions, thereby covering less on the design fundamentals of components. This study seeks to address this gap by providing a comprehensive and coherent review on component-wise design choices in diffusion models. Specifically, we organize this review according to their three key components, namely the forward process, the reverse process, and the sampling procedure. This allows us to provide a fine-grained perspective of diffusion models, benefiting future studies in the analysis of individual components, the applicability of design choices, and the implementation of diffusion models.
☆ Multimodal Learning Without Labeled Multimodal Data: Guarantees and Applications
In many machine learning systems that jointly learn from multiple modalities, a core research question is to understand the nature of multimodal interactions: the emergence of new task-relevant information during learning from both modalities that was not present in either alone. We study this challenge of interaction quantification in a semi-supervised setting with only labeled unimodal data and naturally co-occurring multimodal data (e.g., unlabeled images and captions, video and corresponding audio) but when labeling them is time-consuming. Using a precise information-theoretic definition of interactions, our key contributions are the derivations of lower and upper bounds to quantify the amount of multimodal interactions in this semi-supervised setting. We propose two lower bounds based on the amount of shared information between modalities and the disagreement between separately trained unimodal classifiers, and derive an upper bound through connections to approximate algorithms for min-entropy couplings. We validate these estimated bounds and show how they accurately track true interactions. Finally, two semi-supervised multimodal applications are explored based on these theoretical results: (1) analyzing the relationship between multimodal performance and estimated interactions, and (2) self-supervised learning that embraces disagreement between modalities beyond agreement as is typically done.
comment: Code available at: https://github.com/pliang279/PID
☆ Git-Theta: A Git Extension for Collaborative Development of Machine Learning Models
Currently, most machine learning models are trained by centralized teams and are rarely updated. In contrast, open-source software development involves the iterative development of a shared artifact through distributed collaboration using a version control system. In the interest of enabling collaborative and continual improvement of machine learning models, we introduce Git-Theta, a version control system for machine learning models. Git-Theta is an extension to Git, the most widely used version control software, that allows fine-grained tracking of changes to model parameters alongside code and other artifacts. Unlike existing version control systems that treat a model checkpoint as a blob of data, Git-Theta leverages the structure of checkpoints to support communication-efficient updates, automatic model merges, and meaningful reporting about the difference between two versions of a model. In addition, Git-Theta includes a plug-in system that enables users to easily add support for new functionality. In this paper, we introduce Git-Theta's design and features and include an example use-case of Git-Theta where a pre-trained model is continually adapted and modified. We publicly release Git-Theta in hopes of kickstarting a new era of collaborative model development.
PromptBench: Towards Evaluating the Robustness of Large Language Models on Adversarial Prompts
The increasing reliance on Large Language Models (LLMs) across academia and industry necessitates a comprehensive understanding of their robustness to prompts. In response to this vital need, we introduce PromptBench, a robustness benchmark designed to measure LLMs' resilience to adversarial prompts. This study uses a plethora of adversarial textual attacks targeting prompts across multiple levels: character, word, sentence, and semantic. These prompts are then employed in diverse tasks, such as sentiment analysis, natural language inference, reading comprehension, machine translation, and math problem-solving. Our study generates 4,032 adversarial prompts, meticulously evaluated over 8 tasks and 13 datasets, with 567,084 test samples in total. Our findings demonstrate that contemporary LLMs are vulnerable to adversarial prompts. Furthermore, we present comprehensive analysis to understand the mystery behind prompt robustness and its transferability. We then offer insightful robustness analysis and pragmatic recommendations for prompt composition, beneficial to both researchers and everyday users. We make our code, prompts, and methodologies to generate adversarial prompts publicly accessible, thereby enabling and encouraging collaborative exploration in this pivotal field: https://github.com/microsoft/promptbench.
comment: Technical report; 23 pages
☆ ContriMix: Unsupervised disentanglement of content and attribute for domain generalization in microscopy image analysis
Domain generalization is critical for real-world applications of machine learning models to microscopy images, including histopathology and fluorescence imaging. Artifacts in histopathology arise through a complex combination of factors relating to tissue collection and laboratory processing, as well as factors intrinsic to patient samples. In fluorescence imaging, these artifacts stem from variations across experimental batches. The complexity and subtlety of these artifacts make the enumeration of data domains intractable. Therefore, augmentation-based methods of domain generalization that require domain identifiers and manual fine-tuning are inadequate in this setting. To overcome this challenge, we introduce ContriMix, a domain generalization technique that learns to generate synthetic images by disentangling and permuting the biological content ("content") and technical variations ("attributes") in microscopy images. ContriMix does not rely on domain identifiers or handcrafted augmentations and makes no assumptions about the input characteristics of images. We assess the performance of ContriMix on two pathology datasets (Camelyon17-WILDS and a prostate cell classification dataset) and one fluorescence microscopy dataset (RxRx1-WILDS). ContriMix outperforms current state-of-the-art methods in all datasets, motivating its usage for microscopy image analysis in real-world settings where domain information is hard to come by.
☆ Estimating Koopman operators with sketching to provably learn large scale dynamical systems
The theory of Koopman operators allows to deploy non-parametric machine learning algorithms to predict and analyze complex dynamical systems. Estimators such as principal component regression (PCR) or reduced rank regression (RRR) in kernel spaces can be shown to provably learn Koopman operators from finite empirical observations of the system's time evolution. Scaling these approaches to very long trajectories is a challenge and requires introducing suitable approximations to make computations feasible. In this paper, we boost the efficiency of different kernel-based Koopman operator estimators using random projections (sketching). We derive, implement and test the new "sketched" estimators with extensive experiments on synthetic and large-scale molecular dynamics datasets. Further, we establish non asymptotic error bounds giving a sharp characterization of the trade-offs between statistical learning rates and computational efficiency. Our empirical and theoretical analysis shows that the proposed estimators provide a sound and efficient way to learn large scale dynamical systems. In particular our experiments indicate that the proposed estimators retain the same accuracy of PCR or RRR, while being much faster.
comment: 9 pages, 4 figures
☆ Sample-Level Weighting for Multi-Task Learning with Auxiliary Tasks
Multi-task learning (MTL) can improve the generalization performance of neural networks by sharing representations with related tasks. Nonetheless, MTL can also degrade performance through harmful interference between tasks. Recent work has pursued task-specific loss weighting as a solution for this interference. However, existing algorithms treat tasks as atomic, lacking the ability to explicitly separate harmful and helpful signals beyond the task level. To this end, we propose SLGrad, a sample-level weighting algorithm for multi-task learning with auxiliary tasks. Through sample-specific task weights, SLGrad reshapes the task distributions during training to eliminate harmful auxiliary signals and augment useful task signals. Substantial generalization performance gains are observed on (semi-) synthetic datasets and common supervised multi-task problems.
comment: 16 pages, 7 figures
☆ Optimal sensor placement for reconstructing wind pressure field around buildings using compressed sensing
Deciding how to optimally deploy sensors in a large, complex, and spatially extended structure is critical to ensure that the surface pressure field is accurately captured for subsequent analysis and design. In some cases, reconstruction of missing data is required in downstream tasks such as the development of digital twins. This paper presents a data-driven sparse sensor selection algorithm, aiming to provide the most information contents for reconstructing aerodynamic characteristics of wind pressures over tall building structures parsimoniously. The algorithm first fits a set of basis functions to the training data, then applies a computationally efficient QR algorithm that ranks existing pressure sensors in order of importance based on the state reconstruction to this tailored basis. The findings of this study show that the proposed algorithm successfully reconstructs the aerodynamic characteristics of tall buildings from sparse measurement locations, generating stable and optimal solutions across a range of conditions. As a result, this study serves as a promising first step toward leveraging the success of data-driven and machine learning algorithms to supplement traditional genetic algorithms currently used in wind engineering.
comment: 31 pages, 19 figures
☆ Improving neural network representations using human similarity judgments
Deep neural networks have reached human-level performance on many computer vision tasks. However, the objectives used to train these networks enforce only that similar images are embedded at similar locations in the representation space, and do not directly constrain the global structure of the resulting space. Here, we explore the impact of supervising this global structure by linearly aligning it with human similarity judgments. We find that a naive approach leads to large changes in local representational structure that harm downstream performance. Thus, we propose a novel method that aligns the global structure of representations while preserving their local structure. This global-local transform considerably improves accuracy across a variety of few-shot learning and anomaly detection tasks. Our results indicate that human visual representations are globally organized in a way that facilitates learning from few examples, and incorporating this global structure into neural network representations improves performance on downstream tasks.
☆ Hardness of Deceptive Certificate Selection
Recent progress towards theoretical interpretability guarantees for AI has been made with classifiers that are based on interactive proof systems. A prover selects a certificate from the datapoint and sends it to a verifier who decides the class. In the context of machine learning, such a certificate can be a feature that is informative of the class. For a setup with high soundness and completeness, the exchanged certificates must have a high mutual information with the true class of the datapoint. However, this guarantee relies on a bound on the Asymmetric Feature Correlation of the dataset, a property that so far is difficult to estimate for high-dimensional data. It was conjectured in W\"aldchen et al. that it is computationally hard to exploit the AFC, which is what we prove here. We consider a malicious prover-verifier duo that aims to exploit the AFC to achieve high completeness and soundness while using uninformative certificates. We show that this task is $\mathsf{NP}$-hard and cannot be approximated better than $\mathcal{O}(m^{1/8 - \epsilon})$, where $m$ is the number of possible certificates, for $\epsilon>0$ under the Dense-vs-Random conjecture. This is some evidence that AFC should not prevent the use of interactive classification for real-world tasks, as it is computationally hard to be exploited.
comment: 15 pages, 3 figures
☆ Evaluation of ChatGPT on Biomedical Tasks: A Zero-Shot Comparison with Fine-Tuned Generative Transformers ACL 2023
ChatGPT is a large language model developed by OpenAI. Despite its impressive performance across various tasks, no prior work has investigated its capability in the biomedical domain yet. To this end, this paper aims to evaluate the performance of ChatGPT on various benchmark biomedical tasks, such as relation extraction, document classification, question answering, and summarization. To the best of our knowledge, this is the first work that conducts an extensive evaluation of ChatGPT in the biomedical domain. Interestingly, we find based on our evaluation that in biomedical datasets that have smaller training sets, zero-shot ChatGPT even outperforms the state-of-the-art fine-tuned generative transformer models, such as BioGPT and BioBART. This suggests that ChatGPT's pre-training on large text corpora makes it quite specialized even in the biomedical domain. Our findings demonstrate that ChatGPT has the potential to be a valuable tool for various tasks in the biomedical domain that lack large annotated data.
comment: Accepted by BioNLP@ACL 2023
☆ Learning with Noisy Labels by Adaptive Gradient-Based Outlier Removal ECML
An accurate and substantial dataset is necessary to train a reliable and well-performing model. However, even manually labeled datasets contain errors, not to mention automatically labeled ones. The problem of data denoising was addressed in different existing research, most of which focuses on the detection of outliers and their permanent removal - a process that is likely to over- or underfilter the dataset. In this work, we propose AGRA: a new method for Adaptive GRAdient-based outlier removal. Instead of cleaning the dataset prior to model training, the dataset is adjusted during the training process. By comparing the aggregated gradient of a batch of samples and an individual example gradient, our method dynamically decides whether a corresponding example is helpful for the model at this point or is counter-productive and should be left out for the current update. Extensive evaluation on several datasets demonstrates the AGRA effectiveness, while comprehensive results analysis supports our initial hypothesis: permanent hard outlier removal is not always what model benefits the most from.
comment: Accepted for ECML PKDD 2023
☆ Optimal Fair Multi-Agent Bandits
In this paper, we study the problem of fair multi-agent multi-arm bandit learning when agents do not communicate with each other, except collision information, provided to agents accessing the same arm simultaneously. We provide an algorithm with regret $O\left(N^3 \log N \log T \right)$ (assuming bounded rewards, with unknown bound). This significantly improves previous results which had regret of order $O(\log T \log\log T)$ and exponential dependence on the number of agents. The result is attained by using a distributed auction algorithm to learn the sample-optimal matching, a new type of exploitation phase whose length is derived from the observed samples, and a novel order-statistics-based regret analysis. Simulation results present the dependence of the regret on $\log T$.
comment: 17 pages, 3 figures
☆ Limits, approximation and size transferability for GNNs on sparse graphs via graphops NeurIPS 2023
Can graph neural networks generalize to graphs that are different from the graphs they were trained on, e.g., in size? In this work, we study this question from a theoretical perspective. While recent work established such transferability and approximation results via graph limits, e.g., via graphons, these only apply non-trivially to dense graphs. To include frequently encountered sparse graphs such as bounded-degree or power law graphs, we take a perspective of taking limits of operators derived from graphs, such as the aggregation operation that makes up GNNs. This leads to the recently introduced limit notion of graphops (Backhausz and Szegedy, 2022). We demonstrate how the operator perspective allows us to develop quantitative bounds on the distance between a finite GNN and its limit on an infinite graph, as well as the distance between the GNN on graphs of different sizes that share structural properties, under a regularity assumption verified for various graph sequences. Our results hold for dense and sparse graphs, and various notions of graph limits.
comment: NeurIPS 2023 submission, 34 pages
☆ Fair Column Subset Selection
We consider the problem of fair column subset selection. In particular, we assume that two groups are present in the data, and the chosen column subset must provide a good approximation for both, relative to their respective best rank-k approximations. We show that this fair setting introduces significant challenges: in order to extend known results, one cannot do better than the trivial solution of simply picking twice as many columns as the original methods. We adopt a known approach based on deterministic leverage-score sampling, and show that merely sampling a subset of appropriate size becomes NP-hard in the presence of two groups. Whereas finding a subset of two times the desired size is trivial, we provide an efficient algorithm that achieves the same guarantees with essentially 1.5 times that size. We validate our methods through an extensive set of experiments on real-world data.
☆ Rewarded soups: towards Pareto-optimal alignment by interpolating weights fine-tuned on diverse rewards
Foundation models are first pre-trained on vast unsupervised datasets and then fine-tuned on labeled data. Reinforcement learning, notably from human feedback (RLHF), can further align the network with the intended usage. Yet the imperfections in the proxy reward may hinder the training and lead to suboptimal results; the diversity of objectives in real-world tasks and human opinions exacerbate the issue. This paper proposes embracing the heterogeneity of diverse rewards by following a multi-policy strategy. Rather than focusing on a single a priori reward, we aim for Pareto-optimal generalization across the entire space of preferences. To this end, we propose rewarded soup, first specializing multiple networks independently (one for each proxy reward) and then interpolating their weights linearly. This succeeds empirically because we show that the weights remain linearly connected when fine-tuned on diverse rewards from a shared pre-trained initialization. We demonstrate the effectiveness of our approach for text-to-text (summarization, Q&A, helpful assistant, review), text-image (image captioning, text-to-image generation, visual grounding, VQA), and control (locomotion) tasks. We hope to enhance the alignment of deep models, and how they interact with the world in all its diversity.
☆ Training-Free Neural Active Learning with Initialization-Robustness Guarantees ICML 2023
Existing neural active learning algorithms have aimed to optimize the predictive performance of neural networks (NNs) by selecting data for labelling. However, other than a good predictive performance, being robust against random parameter initializations is also a crucial requirement in safety-critical applications. To this end, we introduce our expected variance with Gaussian processes (EV-GP) criterion for neural active learning, which is theoretically guaranteed to select data points which lead to trained NNs with both (a) good predictive performances and (b) initialization robustness. Importantly, our EV-GP criterion is training-free, i.e., it does not require any training of the NN during data selection, which makes it computationally efficient. We empirically demonstrate that our EV-GP criterion is highly correlated with both initialization robustness and generalization performance, and show that it consistently outperforms baseline methods in terms of both desiderata, especially in situations with limited initial data or large batch sizes.
comment: Accepted to 40th International Conference on Machine Learning (ICML 2023), 41 pages
☆ Multi-modal Latent Diffusion
Multi-modal data-sets are ubiquitous in modern applications, and multi-modal Variational Autoencoders are a popular family of models that aim to learn a joint representation of the different modalities. However, existing approaches suffer from a coherence-quality tradeoff, where models with good generation quality lack generative coherence across modalities, and vice versa. We discuss the limitations underlying the unsatisfactory performance of existing methods, to motivate the need for a different approach. We propose a novel method that uses a set of independently trained, uni-modal, deterministic autoencoders. Individual latent variables are concatenated into a common latent space, which is fed to a masked diffusion model to enable generative modeling. We also introduce a new multi-time training method to learn the conditional score network for multi-modal diffusion. Our methodology substantially outperforms competitors in both generation quality and coherence, as shown through an extensive experimental campaign.
☆ Fast Optimal Locally Private Mean Estimation via Random Projections
We study the problem of locally private mean estimation of high-dimensional vectors in the Euclidean ball. Existing algorithms for this problem either incur sub-optimal error or have high communication and/or run-time complexity. We propose a new algorithmic framework, ProjUnit, for private mean estimation that yields algorithms that are computationally efficient, have low communication complexity, and incur optimal error up to a $1+o(1)$-factor. Our framework is deceptively simple: each randomizer projects its input to a random low-dimensional subspace, normalizes the result, and then runs an optimal algorithm such as PrivUnitG in the lower-dimensional space. In addition, we show that, by appropriately correlating the random projection matrices across devices, we can achieve fast server run-time. We mathematically analyze the error of the algorithm in terms of properties of the random projections, and study two instantiations. Lastly, our experiments for private mean estimation and private federated learning demonstrate that our algorithms empirically obtain nearly the same utility as optimal ones while having significantly lower communication and computational cost.
☆ Dual policy as self-model for planning
Planning is a data efficient decision-making strategy where an agent selects candidate actions by exploring possible future states. To simulate future states when there is a high-dimensional action space, the knowledge of one's decision making strategy must be used to limit the number of actions to be explored. We refer to the model used to simulate one's decisions as the agent's self-model. While self-models are implicitly used widely in conjunction with world models to plan actions, it remains unclear how self-models should be designed. Inspired by current reinforcement learning approaches and neuroscience, we explore the benefits and limitations of using a distilled policy network as the self-model. In such dual-policy agents, a model-free policy and a distilled policy are used for model-free actions and planned actions, respectively. Our results on a ecologically relevant, parametric environment indicate that distilled policy network for self-model stabilizes training, has faster inference than using model-free policy, promotes better exploration, and could learn a comprehensive understanding of its own behaviors, at the cost of distilling a new network apart from the model-free policy.
☆ Faithful Knowledge Distillation
Knowledge distillation (KD) has received much attention due to its success in compressing networks to allow for their deployment in resource-constrained systems. While the problem of adversarial robustness has been studied before in the KD setting, previous works overlook what we term the relative calibration of the student network with respect to its teacher in terms of soft confidences. In particular, we focus on two crucial questions with regard to a teacher-student pair: (i) do the teacher and student disagree at points close to correctly classified dataset examples, and (ii) is the distilled student as confident as the teacher around dataset examples? These are critical questions when considering the deployment of a smaller student network trained from a robust teacher within a safety-critical setting. To address these questions, we introduce a faithful imitation framework to discuss the relative calibration of confidences, as well as provide empirical and certified methods to evaluate the relative calibration of a student w.r.t. its teacher. Further, to verifiably align the relative calibration incentives of the student to those of its teacher, we introduce faithful distillation. Our experiments on the MNIST and Fashion-MNIST datasets demonstrate the need for such an analysis and the advantages of the increased verifiability of faithful distillation over alternative adversarial distillation methods.
comment: 12pgs (main content), 3 figures
☆ Balancing of competitive two-player Game Levels with Reinforcement Learning
The balancing process for game levels in a competitive two-player context involves a lot of manual work and testing, particularly in non-symmetrical game levels. In this paper, we propose an architecture for automated balancing of tile-based levels within the recently introduced PCGRL framework (procedural content generation via reinforcement learning). Our architecture is divided into three parts: (1) a level generator, (2) a balancing agent and, (3) a reward modeling simulation. By playing the level in a simulation repeatedly, the balancing agent is rewarded for modifying it towards the same win rates for all players. To this end, we introduce a novel family of swap-based representations to increase robustness towards playability. We show that this approach is capable to teach an agent how to alter a level for balancing better and faster than plain PCGRL. In addition, by analyzing the agent's swapping behavior, we can draw conclusions about which tile types influence the balancing most. We test and show our results using the Neural MMO (NMMO) environment in a competitive two-player setting.
comment: 8 pages, 8 figures, 1 table. Accepted at IEEE Conference on Games 2023
☆ Towards High-Performance Exploratory Data Analysis (EDA) Via Stable Equilibrium Point
Exploratory data analysis (EDA) is a vital procedure for data science projects. In this work, we introduce a stable equilibrium point (SEP) - based framework for improving the efficiency and solution quality of EDA. By exploiting the SEPs to be the representative points, our approach aims to generate high-quality clustering and data visualization for large-scale data sets. A very unique property of the proposed method is that the SEPs will directly encode the clustering properties of data sets. Compared with prior state-of-the-art clustering and data visualization methods, the proposed methods allow substantially improving computing efficiency and solution quality for large-scale data analysis tasks.
☆ On Computing Optimal Tree Ensembles ICML 2023
Random forests and, more generally, (decision\nobreakdash-)tree ensembles are widely used methods for classification and regression. Recent algorithmic advances allow to compute decision trees that are optimal for various measures such as their size or depth. We are not aware of such research for tree ensembles and aim to contribute to this area. Mainly, we provide two novel algorithms and corresponding lower bounds. First, we are able to carry over and substantially improve on tractability results for decision trees, obtaining a $(6\delta D S)^S \cdot poly$-time algorithm, where $S$ is the number of cuts in the tree ensemble, $D$ the largest domain size, and $\delta$ is the largest number of features in which two examples differ. To achieve this, we introduce the witness-tree technique which also seems promising for practice. Second, we show that dynamic programming, which has been successful for decision trees, may also be viable for tree ensembles, providing an $\ell^n \cdot poly$-time algorithm, where $\ell$ is the number of trees and $n$ the number of examples. Finally, we compare the number of cuts necessary to classify training data sets for decision trees and tree ensembles, showing that ensembles may need exponentially fewer cuts for increasing number of trees.
comment: Accepted at ICML 2023
☆ Generalized Teacher Forcing for Learning Chaotic Dynamics ICML 2023
Chaotic dynamical systems (DS) are ubiquitous in nature and society. Often we are interested in reconstructing such systems from observed time series for prediction or mechanistic insight, where by reconstruction we mean learning geometrical and invariant temporal properties of the system in question (like attractors). However, training reconstruction algorithms like recurrent neural networks (RNNs) on such systems by gradient-descent based techniques faces severe challenges. This is mainly due to exploding gradients caused by the exponential divergence of trajectories in chaotic systems. Moreover, for (scientific) interpretability we wish to have as low dimensional reconstructions as possible, preferably in a model which is mathematically tractable. Here we report that a surprisingly simple modification of teacher forcing leads to provably strictly all-time bounded gradients in training on chaotic systems, and, when paired with a simple architectural rearrangement of a tractable RNN design, piecewise-linear RNNs (PLRNNs), allows for faithful reconstruction in spaces of at most the dimensionality of the observed system. We show on several DS that with these amendments we can reconstruct DS better than current SOTA algorithms, in much lower dimensions. Performance differences were particularly compelling on real world data with which most other methods severely struggled. This work thus led to a simple yet powerful DS reconstruction algorithm which is highly interpretable at the same time.
comment: To be published in the Proceedings of the 40th International Conference on Machine Learning (ICML 2023)
☆ Policy-Based Self-Competition for Planning Problems
AlphaZero-type algorithms may stop improving on single-player tasks in case the value network guiding the tree search is unable to approximate the outcome of an episode sufficiently well. One technique to address this problem is transforming the single-player task through self-competition. The main idea is to compute a scalar baseline from the agent's historical performances and to reshape an episode's reward into a binary output, indicating whether the baseline has been exceeded or not. However, this baseline only carries limited information for the agent about strategies how to improve. We leverage the idea of self-competition and directly incorporate a historical policy into the planning process instead of its scalar performance. Based on the recently introduced Gumbel AlphaZero (GAZ), we propose our algorithm GAZ 'Play-to-Plan' (GAZ PTP), in which the agent learns to find strong trajectories by planning against possible strategies of its past self. We show the effectiveness of our approach in two well-known combinatorial optimization problems, the Traveling Salesman Problem and the Job-Shop Scheduling Problem. With only half of the simulation budget for search, GAZ PTP consistently outperforms all selected single-player variants of GAZ.
comment: 24 pages, 5 figures
☆ A Fair Classifier Embracing Triplet Collapse
In this paper, we study the behaviour of the triplet loss and show that it can be exploited to limit the biases created and perpetuated by machine learning models. Our fair classifier uses the collapse of the triplet loss when its margin is greater than the maximum distance between two points in the latent space, in the case of stochastic triplet selection.
comment: 9 pages, 7 figures, CAp2023
☆ Improving Diffusion-based Image Translation using Asymmetric Gradient Guidance
Diffusion models have shown significant progress in image translation tasks recently. However, due to their stochastic nature, there's often a trade-off between style transformation and content preservation. Current strategies aim to disentangle style and content, preserving the source image's structure while successfully transitioning from a source to a target domain under text or one-shot image conditions. Yet, these methods often require computationally intense fine-tuning of diffusion models or additional neural networks. To address these challenges, here we present an approach that guides the reverse process of diffusion sampling by applying asymmetric gradient guidance. This results in quicker and more stable image manipulation for both text-guided and image-guided image translation. Our model's adaptability allows it to be implemented with both image- and latent-diffusion models. Experiments show that our method outperforms various state-of-the-art models in image translation tasks.
☆ Multilingual Clinical NER: Translation or Cross-lingual Transfer?
Natural language tasks like Named Entity Recognition (NER) in the clinical domain on non-English texts can be very time-consuming and expensive due to the lack of annotated data. Cross-lingual transfer (CLT) is a way to circumvent this issue thanks to the ability of multilingual large language models to be fine-tuned on a specific task in one language and to provide high accuracy for the same task in another language. However, other methods leveraging translation models can be used to perform NER without annotated data in the target language, by either translating the training set or test set. This paper compares cross-lingual transfer with these two alternative methods, to perform clinical NER in French and in German without any training data in those languages. To this end, we release MedNERF a medical NER test set extracted from French drug prescriptions and annotated with the same guidelines as an English dataset. Through extensive experiments on this dataset and on a German medical dataset (Frei and Kramer, 2021), we show that translation-based methods can achieve similar performance to CLT but require more care in their design. And while they can take advantage of monolingual clinical language models, those do not guarantee better results than large general-purpose multilingual models, whether with cross-lingual transfer or translation.
comment: 23 pages, Proceedings of the 5th Clinical Natural Language Processing Workshop
☆ Get More for Less in Decentralized Learning Systems
Decentralized learning (DL) systems have been gaining popularity because they avoid raw data sharing by communicating only model parameters, hence preserving data confidentiality. However, the large size of deep neural networks poses a significant challenge for decentralized training, since each node needs to exchange gigabytes of data, overloading the network. In this paper, we address this challenge with JWINS, a communication-efficient and fully decentralized learning system that shares only a subset of parameters through sparsification. JWINS uses wavelet transform to limit the information loss due to sparsification and a randomized communication cut-off that reduces communication usage without damaging the performance of trained models. We demonstrate empirically with 96 DL nodes on non-IID datasets that JWINS can achieve similar accuracies to full-sharing DL while sending up to 64% fewer bytes. Additionally, on low communication budgets, JWINS outperforms the state-of-the-art communication-efficient DL algorithm CHOCO-SGD by up to 4x in terms of network savings and time.
☆ Label Shift Quantification with Robustness Guarantees via Distribution Feature Matching ECML
Quantification learning deals with the task of estimating the target label distribution under label shift. In this paper, we first present a unifying framework, distribution feature matching (DFM), that recovers as particular instances various estimators introduced in previous literature. We derive a general performance bound for DFM procedures, improving in several key aspects upon previous bounds derived in particular cases. We then extend this analysis to study robustness of DFM procedures in the misspecified setting under departure from the exact label shift hypothesis, in particular in the case of contamination of the target by an unknown distribution. These theoretical findings are confirmed by a detailed numerical study on simulated and real-world datasets. We also introduce an efficient, scalable and robust version of kernel-based DFM using the Random Fourier Feature principle.
comment: Accepted at the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML) 2023
☆ Learning via Wasserstein-Based High Probability Generalisation Bounds
Minimising upper bounds on the population risk or the generalisation gap has been widely used in structural risk minimisation (SRM) - this is in particular at the core of PAC-Bayesian learning. Despite its successes and unfailing surge of interest in recent years, a limitation of the PAC-Bayesian framework is that most bounds involve a Kullback-Leibler (KL) divergence term (or its variations), which might exhibit erratic behavior and fail to capture the underlying geometric structure of the learning problem - hence restricting its use in practical applications. As a remedy, recent studies have attempted to replace the KL divergence in the PAC-Bayesian bounds with the Wasserstein distance. Even though these bounds alleviated the aforementioned issues to a certain extent, they either hold in expectation, are for bounded losses, or are nontrivial to minimize in an SRM framework. In this work, we contribute to this line of research and prove novel Wasserstein distance-based PAC-Bayesian generalisation bounds for both batch learning with independent and identically distributed (i.i.d.) data, and online learning with potentially non-i.i.d. data. Contrary to previous art, our bounds are stronger in the sense that (i) they hold with high probability, (ii) they apply to unbounded (potentially heavy-tailed) losses, and (iii) they lead to optimizable training objectives that can be used in SRM. As a result we derive novel Wasserstein-based PAC-Bayesian learning algorithms and we illustrate their empirical advantage on a variety of experiments.
☆ Label Aware Speech Representation Learning For Language Identification
Speech representation learning approaches for non-semantic tasks such as language recognition have either explored supervised embedding extraction methods using a classifier model or self-supervised representation learning approaches using raw data. In this paper, we propose a novel framework of combining self-supervised representation learning with the language label information for the pre-training task. This framework, termed as Label Aware Speech Representation (LASR) learning, uses a triplet based objective function to incorporate language labels along with the self-supervised loss function. The speech representations are further fine-tuned for the downstream task. The language recognition experiments are performed on two public datasets - FLEURS and Dhwani. In these experiments, we illustrate that the proposed LASR framework improves over the state-of-the-art systems on language identification. We also report an analysis of the robustness of LASR approach to noisy/missing labels as well as its application to multi-lingual speech recognition tasks.
comment: Accepted at Interspeech 2023
☆ Efficient Recruitment Strategy for Collaborative Mobile Crowd Sensing Based on GCN Trustworthiness Prediction
Collaborative Mobile Crowd Sensing (CMCS) enhances data quality and coverage by promoting teamwork in task sensing, with worker recruitment representing a complex multi-objective optimization problem. Existing strategies mainly focus on the characteristics of workers themselves, neglecting the asymmetric trust relationships between them, which affects the rationality of task utility evaluation. To address this, this paper first employs the Mini-Batch K-Means clustering algorithm and deploys edge servers to enable efficient distributed worker recruitment. Historical data and task requirements are utilized to obtain workers' ability types and distances. A trust-directed graph in the worker's social network is input into the Graph Convolutional Network (GCN) framework for training, capturing asymmetric trustworthiness between worker pairs. Privacy leakage is prevented in CMCS scenarios through high trust values between workers. Ultimately, an undirected recruitment graph is constructed using workers' abilities, trust values, and distance weights, transforming the worker recruitment problem into a Maximum Weight Average Subgraph Problem (MWASP). A Tabu Search Recruitment (TSR) algorithm is proposed to rationally recruit a balanced multi-objective optimal task utility worker set for each task. Extensive simulation experiments on four real-world datasets demonstrate the effectiveness of the proposed strategy, outperforming other strategies.
☆ Edge conductivity in PtSe$_2$ nanostructures
PtSe$_2$ is a promising 2D material for nanoelectromechanical sensing and photodetection in the infrared regime. One of its most compelling features is the facile synthesis at temperatures below 500 {\deg}C, which is compatible with current back-end-of-line semiconductor processing. However, this process generates polycrystalline thin films with nanoflake-like domains of 5 to 100 nm size. To investigate the lateral quantum confinement effect in this size regime, we train a deep neural network to obtain an interatomic potential at DFT accuracy and use that to model ribbons, surfaces, nanoflakes, and nanoplatelets of PtSe$_2$ with lateral widths between 5 to 15 nm. We determine which edge terminations are the most stable and find evidence that the electrical conductivity is localized on the edges for lateral sizes below 10 nm. This suggests that the transport channels in thin films of PtSe$_2$ might be dominated by networks of edges, instead of transport through the layers themselves.
☆ Bayesian Optimisation Against Climate Change: Applications and Benchmarks
Bayesian optimisation is a powerful method for optimising black-box functions, popular in settings where the true function is expensive to evaluate and no gradient information is available. Bayesian optimisation can improve responses to many optimisation problems within climate change for which simulator models are unavailable or expensive to sample from. While there have been several feasibility demonstrations of Bayesian optimisation in climate-related applications, there has been no unifying review of applications and benchmarks. We provide such a review here, to encourage the use of Bayesian optimisation in important and well-suited application domains. We identify four main application domains: material discovery, wind farm layout, optimal renewable control and environmental monitoring. For each domain we identify a public benchmark or data set that is easy to use and evaluate systems against, while being representative of real-world problems. Due to the lack of a suitable benchmark for environmental monitoring, we propose LAQN-BO, based on air pollution data. Our contributions are: a) identifying a representative range of benchmarks, providing example code where necessary; b) introducing a new benchmark, LAQN-BO; and c) promoting a wider use of climate change applications among Bayesian optimisation practitioners.
☆ Unpaired Deep Learning for Pharmacokinetic Parameter Estimation from Dynamic Contrast-Enhanced MRI
DCE-MRI provides information about vascular permeability and tissue perfusion through the acquisition of pharmacokinetic parameters. However, traditional methods for estimating these pharmacokinetic parameters involve fitting tracer kinetic models, which often suffer from computational complexity and low accuracy due to noisy arterial input function (AIF) measurements. Although some deep learning approaches have been proposed to tackle these challenges, most existing methods rely on supervised learning that requires paired input DCE-MRI and labeled pharmacokinetic parameter maps. This dependency on labeled data introduces significant time and resource constraints, as well as potential noise in the labels, making supervised learning methods often impractical. To address these limitations, here we present a novel unpaired deep learning method for estimating both pharmacokinetic parameters and the AIF using a physics-driven CycleGAN approach. Our proposed CycleGAN framework is designed based on the underlying physics model, resulting in a simpler architecture with a single generator and discriminator pair. Crucially, our experimental results indicate that our method, which does not necessitate separate AIF measurements, produces more reliable pharmacokinetic parameters than other techniques.
☆ Changing Data Sources in the Age of Machine Learning for Official Statistics
Data science has become increasingly essential for the production of official statistics, as it enables the automated collection, processing, and analysis of large amounts of data. With such data science practices in place, it enables more timely, more insightful and more flexible reporting. However, the quality and integrity of data-science-driven statistics rely on the accuracy and reliability of the data sources and the machine learning techniques that support them. In particular, changes in data sources are inevitable to occur and pose significant risks that are crucial to address in the context of machine learning for official statistics. This paper gives an overview of the main risks, liabilities, and uncertainties associated with changing data sources in the context of machine learning for official statistics. We provide a checklist of the most prevalent origins and causes of changing data sources; not only on a technical level but also regarding ownership, ethics, regulation, and public perception. Next, we highlight the repercussions of changing data sources on statistical reporting. These include technical effects such as concept drift, bias, availability, validity, accuracy and completeness, but also the neutrality and potential discontinuation of the statistical offering. We offer a few important precautionary measures, such as enhancing robustness in both data sourcing and statistical techniques, and thorough monitoring. In doing so, machine learning-based official statistics can maintain integrity, reliability, consistency, and relevance in policy-making, decision-making, and public discourse.
comment: Presented at UNECE Machine Learning for Official Statistics Workshop 2023
☆ CaptAinGlove: Capacitive and Inertial Fusion-Based Glove for Real-Time on Edge Hand Gesture Recognition for Drone Control
We present CaptAinGlove, a textile-based, low-power (1.15Watts), privacy-conscious, real-time on-the-edge (RTE) glove-based solution with a tiny memory footprint (2MB), designed to recognize hand gestures used for drone control. We employ lightweight convolutional neural networks as the backbone models and a hierarchical multimodal fusion to reduce power consumption and improve accuracy. The system yields an F1-score of 80% for the offline evaluation of nine classes; eight hand gesture commands and null activity. For the RTE, we obtained an F1-score of 67% (one user).
☆ Timing Process Interventions with Causal Inference and Reinforcement Learning
The shift from the understanding and prediction of processes to their optimization offers great benefits to businesses and other organizations. Precisely timed process interventions are the cornerstones of effective optimization. Prescriptive process monitoring (PresPM) is the sub-field of process mining that concentrates on process optimization. The emerging PresPM literature identifies state-of-the-art methods, causal inference (CI) and reinforcement learning (RL), without presenting a quantitative comparison. Most experiments are carried out using historical data, causing problems with the accuracy of the methods' evaluations and preempting online RL. Our contribution consists of experiments on timed process interventions with synthetic data that renders genuine online RL and the comparison to CI possible, and allows for an accurate evaluation of the results. Our experiments reveal that RL's policies outperform those from CI and are more robust at the same time. Indeed, the RL policies approach perfect policies. Unlike CI, the unaltered online RL approach can be applied to other, more generic PresPM problems such as next best activity recommendations. Nonetheless, CI has its merits in settings where online learning is not an option.
☆ Phrase Retrieval for Open-Domain Conversational Question Answering with Conversational Dependency Modeling via Contrastive Learning ACL 2023
Open-Domain Conversational Question Answering (ODConvQA) aims at answering questions through a multi-turn conversation based on a retriever-reader pipeline, which retrieves passages and then predicts answers with them. However, such a pipeline approach not only makes the reader vulnerable to the errors propagated from the retriever, but also demands additional effort to develop both the retriever and the reader, which further makes it slower since they are not runnable in parallel. In this work, we propose a method to directly predict answers with a phrase retrieval scheme for a sequence of words, reducing the conventional two distinct subtasks into a single one. Also, for the first time, we study its capability for ODConvQA tasks. However, simply adopting it is largely problematic, due to the dependencies between previous and current turns in a conversation. To address this problem, we further introduce a novel contrastive learning strategy, making sure to reflect previous turns when retrieving the phrase for the current context, by maximizing representational similarities of consecutive turns in a conversation while minimizing irrelevant conversational contexts. We validate our model on two ODConvQA datasets, whose experimental results show that it substantially outperforms the relevant baselines with the retriever-reader. Code is available at: https://github.com/starsuzi/PRO-ConvQA.
comment: Findings of ACL 2023
☆ Revising deep learning methods in parking lot occupancy detection
Parking guidance systems have recently become a popular trend as a part of the smart cities' paradigm of development. The crucial part of such systems is the algorithm allowing drivers to search for available parking lots across regions of interest. The classic approach to this task is based on the application of neural network classifiers to camera records. However, existing systems demonstrate a lack of generalization ability and appropriate testing regarding specific visual conditions. In this study, we extensively evaluate state-of-the-art parking lot occupancy detection algorithms, compare their prediction quality with the recently emerged vision transformers, and propose a new pipeline based on EfficientNet architecture. Performed computational experiments have demonstrated the performance increase in the case of our model, which was evaluated on 5 different datasets.
☆ ColNav: Real-Time Colon Navigation for Colonoscopy
Colorectal cancer screening through colonoscopy continues to be the dominant global standard, as it allows identifying pre-cancerous or adenomatous lesions and provides the ability to remove them during the procedure itself. Nevertheless, failure by the endoscopist to identify such lesions increases the likelihood of lesion progression to subsequent colorectal cancer. Ultimately, colonoscopy remains operator-dependent, and the wide range of quality in colonoscopy examinations among endoscopists is influenced by variations in their technique, training, and diligence. This paper presents a novel real-time navigation guidance system for Optical Colonoscopy (OC). Our proposed system employs a real-time approach that displays both an unfolded representation of the colon and a local indicator directing to un-inspected areas. These visualizations are presented to the physician during the procedure, providing actionable and comprehensible guidance to un-surveyed areas in real-time, while seamlessly integrating into the physician's workflow. Through coverage experimental evaluation, we demonstrated that our system resulted in a higher polyp recall (PR) and high inter-rater reliability with physicians for coverage prediction. These results suggest that our real-time navigation guidance system has the potential to improve the quality and effectiveness of Optical Colonoscopy and ultimately benefit patient outcomes.
☆ Permutaion Equivariant Graph Framelets for Heterophilous Semi-supervised Learning
The nature of heterophilous graphs is significantly different with that of homophilous graphs, which suggests aggregations beyond 1-hop neighborhood and causes difficulties in early graph neural network models. In this paper, we develop a new way to implement multi-scale extraction via constructing Haar-type graph framelets with desired properties of permutation equivariance, efficiency, and sparsity, for deep learning tasks on graphs. We further deisgn a graph framelet neural network model PEGFAN using our constructed graph framelets. The experiments are conducted on a synthetic dataset and 9 benchmark datasets to compare performance with other state-of-the-art models. The result shows that our model can achieve best performance on certain datasets of heterophilous graphs (including the majority of heterophilous datasets with relatively larger sizes and denser connections) and competitive performance on the remaining.
☆ Self-Adjusting Weighted Expected Improvement for Bayesian Optimization
Bayesian Optimization (BO) is a class of surrogate-based, sample-efficient algorithms for optimizing black-box problems with small evaluation budgets. The BO pipeline itself is highly configurable with many different design choices regarding the initial design, surrogate model, and acquisition function (AF). Unfortunately, our understanding of how to select suitable components for a problem at hand is very limited. In this work, we focus on the definition of the AF, whose main purpose is to balance the trade-off between exploring regions with high uncertainty and those with high promise for good solutions. We propose Self-Adjusting Weighted Expected Improvement (SAWEI), where we let the exploration-exploitation trade-off self-adjust in a data-driven manner, based on a convergence criterion for BO. On the noise-free black-box BBOB functions of the COCO benchmarking platform, our method exhibits a favorable any-time performance compared to handcrafted baselines and serves as a robust default choice for any problem structure. The suitability of our method also transfers to HPOBench. With SAWEI, we are a step closer to on-the-fly, data-driven, and robust BO designs that automatically adjust their sampling behavior to the problem at hand.
comment: AutoML Conference 2023
☆ Accounting For Informative Sampling When Learning to Forecast Treatment Outcomes Over Time
Machine learning (ML) holds great potential for accurately forecasting treatment outcomes over time, which could ultimately enable the adoption of more individualized treatment strategies in many practical applications. However, a significant challenge that has been largely overlooked by the ML literature on this topic is the presence of informative sampling in observational data. When instances are observed irregularly over time, sampling times are typically not random, but rather informative -- depending on the instance's characteristics, past outcomes, and administered treatments. In this work, we formalize informative sampling as a covariate shift problem and show that it can prohibit accurate estimation of treatment outcomes if not properly accounted for. To overcome this challenge, we present a general framework for learning treatment outcomes in the presence of informative sampling using inverse intensity-weighting, and propose a novel method, TESAR-CDE, that instantiates this framework using Neural CDEs. Using a simulation environment based on a clinical use case, we demonstrate the effectiveness of our approach in learning under informative sampling.
comment: To appear in the Proceedings of the 40th International Conference on Machine Learning, Honolulu, Hawaii, USA. PMLR 202, 2023
☆ Adversarial Sample Detection Through Neural Network Transport Dynamics ECML
We propose a detector of adversarial samples that is based on the view of neural networks as discrete dynamic systems. The detector tells clean inputs from abnormal ones by comparing the discrete vector fields they follow through the layers. We also show that regularizing this vector field during training makes the network more regular on the data distribution's support, thus making the activations of clean inputs more distinguishable from those of abnormal ones. Experimentally, we compare our detector favorably to other detectors on seen and unseen attacks, and show that the regularization of the network's dynamics improves the performance of adversarial detectors that use the internal embeddings as inputs, while also improving test accuracy.
comment: ECML PKDD 2023
☆ Stochastic Collapse: How Gradient Noise Attracts SGD Dynamics Towards Simpler Subnetworks
In this work, we reveal a strong implicit bias of stochastic gradient descent (SGD) that drives overly expressive networks to much simpler subnetworks, thereby dramatically reducing the number of independent parameters, and improving generalization. To reveal this bias, we identify invariant sets, or subsets of parameter space that remain unmodified by SGD. We focus on two classes of invariant sets that correspond to simpler subnetworks and commonly appear in modern architectures. Our analysis uncovers that SGD exhibits a property of stochastic attractivity towards these simpler invariant sets. We establish a sufficient condition for stochastic attractivity based on a competition between the loss landscape's curvature around the invariant set and the noise introduced by stochastic gradients. Remarkably, we find that an increased level of noise strengthens attractivity, leading to the emergence of attractive invariant sets associated with saddle-points or local maxima of the train loss. We observe empirically the existence of attractive invariant sets in trained deep neural networks, implying that SGD dynamics often collapses to simple subnetworks with either vanishing or redundant neurons. We further demonstrate how this simplifying process of stochastic collapse benefits generalization in a linear teacher-student framework. Finally, through this analysis, we mechanistically explain why early training with large learning rates for extended periods benefits subsequent generalization.
comment: 30 pages, 10 figures
☆ Data Mining for Faster, Interpretable Solutions to Inverse Problems: A Case Study Using Additive Manufacturing
Solving inverse problems, where we find the input values that result in desired values of outputs, can be challenging. The solution process is often computationally expensive and it can be difficult to interpret the solution in high-dimensional input spaces. In this paper, we use a problem from additive manufacturing to address these two issues with the intent of making it easier to solve inverse problems and exploit their results. First, focusing on Gaussian process surrogates that are used to solve inverse problems, we describe how a simple modification to the idea of tapering can substantially speed up the surrogate without losing accuracy in prediction. Second, we demonstrate that Kohonen self-organizing maps can be used to visualize and interpret the solution to the inverse problem in the high-dimensional input space. For our data set, as not all input dimensions are equally important, we show that using weighted distances results in a better organized map that makes the relationships among the inputs obvious.
comment: 16 figures and 4 tables
☆ Normalization Layers Are All That Sharpness-Aware Minimization Needs
Sharpness-aware minimization (SAM) was proposed to reduce sharpness of minima and has been shown to enhance generalization performance in various settings. In this work we show that perturbing only the affine normalization parameters (comprising less than 0.1% of the total parameters) in the adversarial step of SAM outperforms perturbing all of the parameters. This finding generalizes to different SAM variants and both ResNet (Batch Normalization) and Vision Transformer (Layer Normalization) architectures. We consider alternative sparse perturbation approaches and find that these do not achieve similar performance enhancement at such extreme sparsity levels, showing that this behaviour is unique to the normalization layers. Although our findings reaffirm the effectiveness of SAM in improving generalization performance, they cast doubt on whether this is solely caused by reduced sharpness. The code for our experiments is publicly available at https://github.com/mueller-mp/SAM-ON.
☆ Efficient Vision Transformer for Human Pose Estimation via Patch Selection
While Convolutional Neural Networks (CNNs) have been widely successful in 2D human pose estimation, Vision Transformers (ViTs) have emerged as a promising alternative to CNNs, boosting state-of-the-art performance. However, the quadratic computational complexity of ViTs has limited their applicability for processing high-resolution images and long videos. To address this challenge, we propose a simple method for reducing ViT's computational complexity based on selecting and processing a small number of most informative patches while disregarding others. We leverage a lightweight pose estimation network to guide the patch selection process, ensuring that the selected patches contain the most important information. Our experimental results on three widely used 2D pose estimation benchmarks, namely COCO, MPII and OCHuman, demonstrate the effectiveness of our proposed methods in significantly improving speed and reducing computational complexity with a slight drop in performance.
☆ Causally Learning an Optimal Rework Policy
In manufacturing, rework refers to an optional step of a production process which aims to eliminate errors or remedy products that do not meet the desired quality standards. Reworking a production lot involves repeating a previous production stage with adjustments to ensure that the final product meets the required specifications. While offering the chance to improve the yield and thus increase the revenue of a production lot, a rework step also incurs additional costs. Additionally, the rework of parts that already meet the target specifications may damage them and decrease the yield. In this paper, we apply double/debiased machine learning (DML) to estimate the conditional treatment effect of a rework step during the color conversion process in opto-electronic semiconductor manufacturing on the final product yield. We utilize the implementation DoubleML to develop policies for the rework of components and estimate their value empirically. From our causal machine learning analysis we derive implications for the coating of monochromatic LEDs with conversion layers.
comment: 22 pages, 15 figures
☆ Look Beneath the Surface: Exploiting Fundamental Symmetry for Sample-Efficient Offline RL
Offline reinforcement learning (RL) offers an appealing approach to real-world tasks by learning policies from pre-collected datasets without interacting with the environment. However, the performance of existing offline RL algorithms heavily depends on the scale and state-action space coverage of datasets. Real-world data collection is often expensive and uncontrollable, leading to small and narrowly covered datasets and posing significant challenges for practical deployments of offline RL. In this paper, we provide a new insight that leveraging the fundamental symmetry of system dynamics can substantially enhance offline RL performance under small datasets. Specifically, we propose a Time-reversal symmetry (T-symmetry) enforced Dynamics Model (TDM), which establishes consistency between a pair of forward and reverse latent dynamics. TDM provides both well-behaved representations for small datasets and a new reliability measure for OOD samples based on compliance with the T-symmetry. These can be readily used to construct a new offline RL algorithm (TSRL) with less conservative policy constraints and a reliable latent space data augmentation procedure. Based on extensive experiments, we find TSRL achieves great performance on small benchmark datasets with as few as 1% of the original samples, which significantly outperforms the recent offline RL algorithms in terms of data efficiency and generalizability.
comment: The first two authors contributed equally
☆ DualHGNN: A Dual Hypergraph Neural Network for Semi-Supervised Node Classification based on Multi-View Learning and Density Awareness IJCNN 2023
Graph-based semi-supervised node classification has been shown to become a state-of-the-art approach in many applications with high research value and significance. Most existing methods are only based on the original intrinsic or artificially established graph structure which may not accurately reflect the "true" correlation among data and are not optimal for semi-supervised node classification in the downstream graph neural networks. Besides, while existing graph-based methods mostly utilize the explicit graph structure, some implicit information, for example, the density information, can also provide latent information that can be further exploited. To address these limitations, this paper proposes the Dual Hypergraph Neural Network (DualHGNN), a new dual connection model integrating both hypergraph structure learning and hypergraph representation learning simultaneously in a unified architecture. The DualHGNN first leverages a multi-view hypergraph learning network to explore the optimal hypergraph structure from multiple views, constrained by a consistency loss proposed to improve its generalization. Then, DualHGNN employs a density-aware hypergraph attention network to explore the high-order semantic correlation among data points based on the density-aware attention mechanism. Extensive experiments are conducted in various benchmark datasets, and the results demonstrate the effectiveness of the proposed approach.
comment: This work has been accepted by 2023 International Joint Conference on Neural Networks (IJCNN 2023). arXiv admin note: text overlap with arXiv:2201.11511
☆ Migrate Demographic Group For Fair GNNs
Graph Neural networks (GNNs) have been applied in many scenarios due to the superior performance of graph learning. However, fairness is always ignored when designing GNNs. As a consequence, biased information in training data can easily affect vanilla GNNs, causing biased results toward particular demographic groups (divided by sensitive attributes, such as race and age). There have been efforts to address the fairness issue. However, existing fair techniques generally divide the demographic groups by raw sensitive attributes and assume that are fixed. The biased information correlated with raw sensitive attributes will run through the training process regardless of the implemented fair techniques. It is urgent to resolve this problem for training fair GNNs. To tackle this problem, we propose a brand new framework, FairMigration, which can dynamically migrate the demographic groups instead of keeping that fixed with raw sensitive attributes. FairMigration is composed of two training stages. In the first stage, the GNNs are initially optimized by personalized self-supervised learning, and the demographic groups are adjusted dynamically. In the second stage, the new demographic groups are frozen and supervised learning is carried out under the constraints of new demographic groups and adversarial training. Extensive experiments reveal that FairMigration balances model performance and fairness well.
☆ Leveraging Knowledge Graph Embeddings to Enhance Contextual Representations for Relation Extraction
Relation extraction task is a crucial and challenging aspect of Natural Language Processing. Several methods have surfaced as of late, exhibiting notable performance in addressing the task; however, most of these approaches rely on vast amounts of data from large-scale knowledge graphs or language models pretrained on voluminous corpora. In this paper, we hone in on the effective utilization of solely the knowledge supplied by a corpus to create a high-performing model. Our objective is to showcase that by leveraging the hierarchical structure and relational distribution of entities within a corpus without introducing external knowledge, a relation extraction model can achieve significantly enhanced performance. We therefore proposed a relation extraction approach based on the incorporation of pretrained knowledge graph embeddings at the corpus scale into the sentence-level contextual representation. We conducted a series of experiments which revealed promising and very interesting results for our proposed approach.The obtained results demonstrated an outperformance of our method compared to context-based relation extraction models.
comment: 15 pages, 1 figures, The 17th International Conference on Document Analysis and Recognition
☆ Improving Hyperparameter Learning under Approximate Inference in Gaussian Process Models ICML
Approximate inference in Gaussian process (GP) models with non-conjugate likelihoods gets entangled with the learning of the model hyperparameters. We improve hyperparameter learning in GP models and focus on the interplay between variational inference (VI) and the learning target. While VI's lower bound to the marginal likelihood is a suitable objective for inferring the approximate posterior, we show that a direct approximation of the marginal likelihood as in Expectation Propagation (EP) is a better learning objective for hyperparameter optimization. We design a hybrid training procedure to bring the best of both worlds: it leverages conjugate-computation VI for inference and uses an EP-like marginal likelihood approximation for hyperparameter learning. We compare VI, EP, Laplace approximation, and our proposed training procedure and empirically demonstrate the effectiveness of our proposal across a wide range of data sets.
comment: International Conference on Machine Learning (ICML) 2023
☆ An ASR-Based Tutor for Learning to Read: How to Optimize Feedback to First Graders SP
The interest in employing automatic speech recognition (ASR) in applications for reading practice has been growing in recent years. In a previous study, we presented an ASR-based Dutch reading tutor application that was developed to provide instantaneous feedback to first-graders learning to read. We saw that ASR has potential at this stage of the reading process, as the results suggested that pupils made progress in reading accuracy and fluency by using the software. In the current study, we used children's speech from an existing corpus (JASMIN) to develop two new ASR systems, and compared the results to those of the previous study. We analyze correct/incorrect classification of the ASR systems using human transcripts at word level, by means of evaluation measures such as Cohen's Kappa, Matthews Correlation Coefficient (MCC), precision, recall and F-measures. We observe improvements for the newly developed ASR systems regarding the agreement with human-based judgment and correct rejection (CR). The accuracy of the ASR systems varies for different reading tasks and word types. Our results suggest that, in the current configuration, it is difficult to classify isolated words. We discuss these results, possible ways to improve our systems and avenues for future research.
comment: Published (double-blind peer-reviewed) on SPECOM 2021
Self-supervised Audio Teacher-Student Transformer for Both Clip-level and Frame-level Tasks
In recent years, self-supervised learning (SSL) has emerged as a popular approach for learning audio representations. The ultimate goal of audio self-supervised pre-training is to transfer knowledge to downstream audio tasks, generally including clip-level and frame-level tasks. Clip-level tasks classify the scene or sound of an entire audio clip, e.g. audio tagging, instrument recognition, etc. While frame-level tasks detect event-level timestamps from an audio clip, e.g. sound event detection, speaker diarization, etc. Prior studies primarily evaluate on clip-level downstream tasks. Frame-level tasks are important for fine-grained acoustic scene/event understanding, and are generally more challenging than clip-level tasks. In order to tackle both clip-level and frame-level tasks, this paper proposes two self-supervised audio representation learning methods: ATST-Clip and ATST-Frame, responsible for learning clip-level and frame-level representations, respectively. ATST stands for Audio Teacher-Student Transformer, which means both methods use a transformer encoder and a teacher-student training scheme.Experimental results show that our ATST-Frame model obtains state-of-the-art (SOTA) performance on most of the clip-level and frame-level downstream tasks. Especially, it outperforms other models by a large margin on the frame-level sound event detection task. In addition, the performance can be further improved by combining the two models through knowledge distillation.
comment: Submitted to IEEE TASLP. arXiv admin note: text overlap with arXiv:2204.12076
☆ Benchmarking Foundation Models with Language-Model-as-an-Examiner
Numerous benchmarks have been established to assess the performance of foundation models on open-ended question answering, which serves as a comprehensive test of a model's ability to understand and generate language in a manner similar to humans. Most of these works focus on proposing new datasets, however, we see two main issues within previous benchmarking pipelines, namely testing leakage and evaluation automation. In this paper, we propose a novel benchmarking framework, Language-Model-as-an-Examiner, where the LM serves as a knowledgeable examiner that formulates questions based on its knowledge and evaluates responses in a reference-free manner. Our framework allows for effortless extensibility as various LMs can be adopted as the examiner, and the questions can be constantly updated given more diverse trigger topics. For a more comprehensive and equitable evaluation, we devise three strategies: (1) We instruct the LM examiner to generate questions across a multitude of domains to probe for a broad acquisition, and raise follow-up questions to engage in a more in-depth assessment. (2) Upon evaluation, the examiner combines both scoring and ranking measurements, providing a reliable result as it aligns closely with human annotations. (3) We additionally propose a decentralized Peer-examination method to address the biases in a single examiner. Our data and benchmarking results are available at: https://lmexam.com.
comment: 23 pages, 8 figures
☆ Optimal Transport Model Distributional Robustness
Distributional robustness is a promising framework for training deep learning models that are less vulnerable to adversarial examples and data distribution shifts. Previous works have mainly focused on exploiting distributional robustness in data space. In this work, we explore an optimal transport-based distributional robustness framework on model spaces. Specifically, we examine a model distribution in a Wasserstein ball of a given center model distribution that maximizes the loss. We have developed theories that allow us to learn the optimal robust center model distribution. Interestingly, through our developed theories, we can flexibly incorporate the concept of sharpness awareness into training a single model, ensemble models, and Bayesian Neural Networks by considering specific forms of the center model distribution, such as a Dirac delta distribution over a single model, a uniform distribution over several models, and a general Bayesian Neural Network. Furthermore, we demonstrate that sharpness-aware minimization (SAM) is a specific case of our framework when using a Dirac delta distribution over a single model, while our framework can be viewed as a probabilistic extension of SAM. We conduct extensive experiments to demonstrate the usefulness of our framework in the aforementioned settings, and the results show remarkable improvements in our approaches to the baselines.
☆ End-to-End Learning for Stochastic Optimization: A Bayesian Perspective ICML 2023
We develop a principled approach to end-to-end learning in stochastic optimization. First, we show that the standard end-to-end learning algorithm admits a Bayesian interpretation and trains a posterior Bayes action map. Building on the insights of this analysis, we then propose new end-to-end learning algorithms for training decision maps that output solutions of empirical risk minimization and distributionally robust optimization problems, two dominant modeling paradigms in optimization under uncertainty. Numerical results for a synthetic newsvendor problem illustrate the key differences between alternative training schemes. We also investigate an economic dispatch problem based on real data to showcase the impact of the neural network architecture of the decision maps on their test performance.
comment: Accepted at ICML 2023
☆ Efficient Alternating Minimization with Applications to Weighted Low Rank Approximation
Weighted low rank approximation is a fundamental problem in numerical linear algebra, and it has many applications in machine learning. Given a matrix $M \in \mathbb{R}^{n \times n}$, a weight matrix $W \in \mathbb{R}_{\geq 0}^{n \times n}$, a parameter $k$, the goal is to output two matrices $U, V \in \mathbb{R}^{n \times k}$ such that $\| W \circ (M - U V) \|_F$ is minimized, where $\circ$ denotes the Hadamard product. Such a problem is known to be NP-hard and even hard to approximate [RSW16]. Meanwhile, alternating minimization is a good heuristic solution for approximating weighted low rank approximation. The work [LLR16] shows that, under mild assumptions, alternating minimization does provide provable guarantees. In this work, we develop an efficient and robust framework for alternating minimization. For weighted low rank approximation, this improves the runtime of [LLR16] from $n^2 k^2$ to $n^2k$. At the heart of our work framework is a high-accuracy multiple response regression solver together with a robust analysis of alternating minimization.
comment: arXiv admin note: text overlap with arXiv:2302.11068
☆ Rethinking Weak Supervision in Helping Contrastive Learning ICML2023
Contrastive learning has shown outstanding performances in both supervised and unsupervised learning, and has recently been introduced to solve weakly supervised learning problems such as semi-supervised learning and noisy label learning. Despite the empirical evidence showing that semi-supervised labels improve the representations of contrastive learning, it remains unknown if noisy supervised information can be directly used in training instead of after manual denoising. Therefore, to explore the mechanical differences between semi-supervised and noisy-labeled information in helping contrastive learning, we establish a unified theoretical framework of contrastive learning under weak supervision. Specifically, we investigate the most intuitive paradigm of jointly training supervised and unsupervised contrastive losses. By translating the weakly supervised information into a similarity graph under the framework of spectral clustering based on the posterior probability of weak labels, we establish the downstream classification error bound. We prove that semi-supervised labels improve the downstream error bound whereas noisy labels have limited effects under such a paradigm. Our theoretical findings here provide new insights for the community to rethink the role of weak supervision in helping contrastive learning.
comment: Accepted to ICML2023
♻ ☆ Balanced Product of Calibrated Experts for Long-Tailed Recognition CVPR 2023
Many real-world recognition problems are characterized by long-tailed label distributions. These distributions make representation learning highly challenging due to limited generalization over the tail classes. If the test distribution differs from the training distribution, e.g. uniform versus long-tailed, the problem of the distribution shift needs to be addressed. A recent line of work proposes learning multiple diverse experts to tackle this issue. Ensemble diversity is encouraged by various techniques, e.g. by specializing different experts in the head and the tail classes. In this work, we take an analytical approach and extend the notion of logit adjustment to ensembles to form a Balanced Product of Experts (BalPoE). BalPoE combines a family of experts with different test-time target distributions, generalizing several previous approaches. We show how to properly define these distributions and combine the experts in order to achieve unbiased predictions, by proving that the ensemble is Fisher-consistent for minimizing the balanced error. Our theoretical analysis shows that our balanced ensemble requires calibrated experts, which we achieve in practice using mixup. We conduct extensive experiments and our method obtains new state-of-the-art results on three long-tailed datasets: CIFAR-100-LT, ImageNet-LT, and iNaturalist-2018. Our code is available at https://github.com/emasa/BalPoE-CalibratedLT.
comment: Accepted at CVPR 2023, 19 pages
♻ ☆ Language Models can Solve Computer Tasks
Agents capable of carrying out general tasks on a computer can improve efficiency and productivity by automating repetitive tasks and assisting in complex problem-solving. Ideally, such agents should be able to solve new computer tasks presented to them through natural language commands. However, previous approaches to this problem require large amounts of expert demonstrations and task-specific reward functions, both of which are impractical for new tasks. In this work, we show that a pre-trained large language model (LLM) agent can execute computer tasks guided by natural language using a simple prompting scheme where the agent Recursively Criticizes and Improves its output (RCI). The RCI approach significantly outperforms existing LLM methods for automating computer tasks and surpasses supervised learning (SL) and reinforcement learning (RL) approaches on the MiniWoB++ benchmark. We compare multiple LLMs and find that RCI with the InstructGPT-3+RLHF LLM is state-of-the-art on MiniWoB++, using only a handful of demonstrations per task rather than tens of thousands, and without a task-specific reward function. Furthermore, we demonstrate RCI prompting's effectiveness in enhancing LLMs' reasoning abilities on a suite of natural language reasoning tasks, outperforming chain of thought (CoT) prompting. We find that RCI combined with CoT performs better than either separately. Our code can be found here: https://github.com/posgnu/rci-agent.
♻ ☆ Smooth Non-Stationary Bandits ICML 2023
In many applications of online decision making, the environment is non-stationary and it is therefore crucial to use bandit algorithms that handle changes. Most existing approaches are designed to protect against non-smooth changes, constrained only by total variation or Lipschitzness over time, where they guarantee $\tilde \Theta(T^{2/3})$ regret. However, in practice environments are often changing {\bf smoothly}, so such algorithms may incur higher-than-necessary regret in these settings and do not leverage information on the rate of change. We study a non-stationary two-armed bandits problem where we assume that an arm's mean reward is a $\beta$-H\"older function over (normalized) time, meaning it is $(\beta-1)$-times Lipschitz-continuously differentiable. We show the first separation between the smooth and non-smooth regimes by presenting a policy with $\tilde O(T^{3/5})$ regret for $\beta=2$. We complement this result by an $\Omg(T^{(\beta+1)/(2\beta+1)})$ lower bound for any integer $\beta\ge 1$, which matches our upper bound for $\beta=2$.
comment: Accepted by ICML 2023
♻ ☆ Mixed Autoencoder for Self-supervised Visual Representation Learning CVPR 2023
Masked Autoencoder (MAE) has demonstrated superior performance on various vision tasks via randomly masking image patches and reconstruction. However, effective data augmentation strategies for MAE still remain open questions, different from those in contrastive learning that serve as the most important part. This paper studies the prevailing mixing augmentation for MAE. We first demonstrate that naive mixing will in contrast degenerate model performance due to the increase of mutual information (MI). To address, we propose homologous recognition, an auxiliary pretext task, not only to alleviate the MI increasement by explicitly requiring each patch to recognize homologous patches, but also to perform object-aware self-supervised pre-training for better downstream dense perception performance. With extensive experiments, we demonstrate that our proposed Mixed Autoencoder (MixedAE) achieves the state-of-the-art transfer results among masked image modeling (MIM) augmentations on different downstream tasks with significant efficiency. Specifically, our MixedAE outperforms MAE by +0.3% accuracy, +1.7 mIoU and +0.9 AP on ImageNet-1K, ADE20K and COCO respectively with a standard ViT-Base. Moreover, MixedAE surpasses iBOT, a strong MIM method combined with instance discrimination, while accelerating training by 2x. To our best knowledge, this is the very first work to consider mixing for MIM from the perspective of pretext task design. Code will be made available.
comment: Accepted by CVPR 2023
♻ ☆ ChatDB: Augmenting LLMs with Databases as Their Symbolic Memory
Large language models (LLMs) with memory are computationally universal. However, mainstream LLMs are not taking full advantage of memory, and the designs are heavily influenced by biological brains. Due to their approximate nature and proneness to the accumulation of errors, conventional neural memory mechanisms cannot support LLMs to simulate complex reasoning. In this paper, we seek inspiration from modern computer architectures to augment LLMs with symbolic memory for complex multi-hop reasoning. Such a symbolic memory framework is instantiated as an LLM and a set of SQL databases, where the LLM generates SQL instructions to manipulate the SQL databases. We validate the effectiveness of the proposed memory framework on a synthetic dataset requiring complex reasoning. The project website is available at https://chatdatabase.github.io/ .
♻ ☆ Gradient boosting for convex cone predict and optimize problems
Prediction models are typically optimized independently from decision optimization. A smart predict then optimize (SPO) framework optimizes prediction models to minimize downstream decision regret. In this paper we present dboost, the first general purpose implementation of smart gradient boosting for `predict, then optimize' problems. The framework supports convex quadratic cone programming and gradient boosting is performed by implicit differentiation of a custom fixed-point mapping. Experiments comparing with state-of-the-art SPO methods show that dboost can further reduce out-of-sample decision regret.
♻ ☆ GAD-NR: Graph Anomaly Detection via Neighborhood Reconstruction
Graph Anomaly Detection (GAD) is a technique used to identify abnormal nodes within graphs, finding applications in network security, fraud detection, social media spam detection, and various other domains. A common method for GAD is Graph Auto-Encoders (GAEs), which encode graph data into node representations and identify anomalies by assessing the reconstruction quality of the graphs based on these representations. However, existing GAE models are primarily optimized for direct link reconstruction, resulting in nodes connected in the graph being clustered in the latent space. As a result, they excel at detecting cluster-type structural anomalies but struggle with more complex structural anomalies that do not conform to clusters. To address this limitation, we propose a novel solution called GAD-NR, a new variant of GAE that incorporates neighborhood reconstruction for graph anomaly detection. GAD-NR aims to reconstruct the entire neighborhood of a node, encompassing the local structure, self-attributes, and neighbor attributes, based on the corresponding node representation. By comparing the neighborhood reconstruction loss between anomalous nodes and normal nodes, GAD-NR can effectively detect any anomalies. Extensive experimentation conducted on six real-world datasets validates the effectiveness of GAD-NR, showcasing significant improvements (by up to 30% in AUC) over state-of-the-art competitors. The source code for GAD-NR is openly available. Importantly, the comparative analysis reveals that the existing methods perform well only in detecting one or two types of anomalies out of the three types studied. In contrast, GAD-NR excels at detecting all three types of anomalies across the datasets, demonstrating its comprehensive anomaly detection capabilities.
♻ ☆ Learning to Suggest Breaks: Sustainable Optimization of Long-Term User Engagement ICML 2023
Optimizing user engagement is a key goal for modern recommendation systems, but blindly pushing users towards increased consumption risks burn-out, churn, or even addictive habits. To promote digital well-being, most platforms now offer a service that periodically prompts users to take breaks. These, however, must be set up manually, and so may be suboptimal for both users and the system. In this paper, we study the role of breaks in recommendation, and propose a framework for learning optimal breaking policies that promote and sustain long-term engagement. Based on the notion that recommendation dynamics are susceptible to both positive and negative feedback, we cast recommendation as a Lotka-Volterra dynamical system, where breaking reduces to a problem of optimal control. We then give an efficient learning algorithm, provide theoretical guarantees, and empirically demonstrate the utility of our approach on semi-synthetic data.
comment: Accepted for publication in ICML 2023
♻ ☆ Meta-learning Control Variates: Variance Reduction with Limited Data UAI 2023
Control variates can be a powerful tool to reduce the variance of Monte Carlo estimators, but constructing effective control variates can be challenging when the number of samples is small. In this paper, we show that when a large number of related integrals need to be computed, it is possible to leverage the similarity between these integration tasks to improve performance even when the number of samples per task is very small. Our approach, called meta learning CVs (Meta-CVs), can be used for up to hundreds or thousands of tasks. Our empirical assessment indicates that Meta-CVs can lead to significant variance reduction in such settings, and our theoretical analysis establishes general conditions under which Meta-CVs can be successfully trained.
comment: Accepted for publication (with an oral presentation) at UAI 2023
♻ ☆ Time-Conditioned Generative Modeling of Object-Centric Representations for Video Decomposition and Prediction
When perceiving the world from multiple viewpoints, humans have the ability to reason about the complete objects in a compositional manner even when an object is completely occluded from certain viewpoints. Meanwhile, humans are able to imagine novel views after observing multiple viewpoints. Recent remarkable advances in multi-view object-centric learning still leaves some unresolved problems: 1) The shapes of partially or completely occluded objects can not be well reconstructed. 2) The novel viewpoint prediction depends on expensive viewpoint annotations rather than implicit rules in view representations. In this paper, we introduce a time-conditioned generative model for videos. To reconstruct the complete shape of an object accurately, we enhance the disentanglement between the latent representations of objects and views, where the latent representations of time-conditioned views are jointly inferred with a Transformer and then are input to a sequential extension of Slot Attention to learn object-centric representations. In addition, Gaussian processes are employed as priors of view latent variables for video generation and novel-view prediction without viewpoint annotations. Experiments on multiple datasets demonstrate that the proposed model can make object-centric video decomposition, reconstruct the complete shapes of occluded objects, and make novel-view predictions.
♻ ☆ A Theory of Link Prediction via Relational Weisfeiler-Leman
Graph neural networks are prominent models for representation learning over graph-structured data. While the capabilities and limitations of these models are well-understood for simple graphs, our understanding remains incomplete in the context of knowledge graphs. Our goal is to provide a systematic understanding of the landscape of graph neural networks for knowledge graphs pertaining to the prominent task of link prediction. Our analysis entails a unifying perspective on seemingly unrelated models and unlocks a series of other models. The expressive power of various models is characterized via a corresponding relational Weisfeiler-Leman algorithm. This analysis is extended to provide a precise logical characterization of the class of functions captured by a class of graph neural networks. The theoretical findings presented in this paper explain the benefits of some widely employed practical design choices, which are validated empirically.
♻ ☆ Extrapolative Controlled Sequence Generation via Iterative Refinement ICML 2023
We study the problem of extrapolative controlled generation, i.e., generating sequences with attribute values beyond the range seen in training. This task is of significant importance in automated design, especially drug discovery, where the goal is to design novel proteins that are \textit{better} (e.g., more stable) than existing sequences. Thus, by definition, the target sequences and their attribute values are out of the training distribution, posing challenges to existing methods that aim to directly generate the target sequence. Instead, in this work, we propose Iterative Controlled Extrapolation (ICE) which iteratively makes local edits to a sequence to enable extrapolation. We train the model on synthetically generated sequence pairs that demonstrate small improvement in the attribute value. Results on one natural language task (sentiment analysis) and two protein engineering tasks (ACE2 stability and AAV fitness) show that ICE considerably outperforms state-of-the-art approaches despite its simplicity. Our code and models are available at: https://github.com/vishakhpk/iter-extrapolation.
comment: ICML 2023 - Camera Ready Version
♻ ☆ A Filtering-based General Approach to Learning Rational Constraints of Epistemic Graphs
Epistemic graphs are a generalization of the epistemic approach to probabilistic argumentation. Hunter proposed a 2-way generalization framework to learn epistemic constraints from crowd-sourcing data. However, the learnt epistemic constraints only reflect users' beliefs from data, without considering the rationality encoded in epistemic graphs. Meanwhile, the current framework can only generate epistemic constraints that reflect whether an agent believes an argument, but not the degree to which it believes in it. The major challenge to achieving this effect is that the computational complexity will increase sharply when expanding the variety of constraints, which may lead to unacceptable time performance. To address these problems, we propose a filtering-based approach using a multiple-way generalization step to generate a set of rational rules which are consistent with their epistemic graphs from a dataset. This approach is able to learn a wider variety of rational rules that reflect information in both the domain model and the user model. Moreover, to improve computational efficiency, we introduce a new function to exclude meaningless rules. The empirical results show that our approach significantly outperforms the existing framework when expanding the variety of rules.
comment: 18 pages, 6 figures, submitted to CLAR 2023
♻ ☆ A Fast, Well-Founded Approximation to the Empirical Neural Tangent Kernel
Empirical neural tangent kernels (eNTKs) can provide a good understanding of a given network's representation: they are often far less expensive to compute and applicable more broadly than infinite width NTKs. For networks with O output units (e.g. an O-class classifier), however, the eNTK on N inputs is of size $NO \times NO$, taking $O((NO)^2)$ memory and up to $O((NO)^3)$ computation. Most existing applications have therefore used one of a handful of approximations yielding $N \times N$ kernel matrices, saving orders of magnitude of computation, but with limited to no justification. We prove that one such approximation, which we call "sum of logits", converges to the true eNTK at initialization for any network with a wide final "readout" layer. Our experiments demonstrate the quality of this approximation for various uses across a range of settings.
♻ ☆ FPUS23: An Ultrasound Fetus Phantom Dataset with Deep Neural Network Evaluations for Fetus Orientations, Fetal Planes, and Anatomical Features
Ultrasound imaging is one of the most prominent technologies to evaluate the growth, progression, and overall health of a fetus during its gestation. However, the interpretation of the data obtained from such studies is best left to expert physicians and technicians who are trained and well-versed in analyzing such images. To improve the clinical workflow and potentially develop an at-home ultrasound-based fetal monitoring platform, we present a novel fetus phantom ultrasound dataset, FPUS23, which can be used to identify (1) the correct diagnostic planes for estimating fetal biometric values, (2) fetus orientation, (3) their anatomical features, and (4) bounding boxes of the fetus phantom anatomies at 23 weeks gestation. The entire dataset is composed of 15,728 images, which are used to train four different Deep Neural Network models, built upon a ResNet34 backbone, for detecting aforementioned fetus features and use-cases. We have also evaluated the models trained using our FPUS23 dataset, to show that the information learned by these models can be used to substantially increase the accuracy on real-world ultrasound fetus datasets. We make the FPUS23 dataset and the pre-trained models publicly accessible at https://github.com/bharathprabakaran/FPUS23, which will further facilitate future research on fetal ultrasound imaging and analysis.
comment: Accepted for Publication at IEEE Access
♻ ☆ Multi-Party Chat: Conversational Agents in Group Settings with Humans and Models
Current dialogue research primarily studies pairwise (two-party) conversations, and does not address the everyday setting where more than two speakers converse together. In this work, we both collect and evaluate multi-party conversations to study this more general case. We use the LIGHT environment to construct grounded conversations, where each participant has an assigned character to role-play. We thus evaluate the ability of language models to act as one or more characters in such conversations. Models require two skills that pairwise-trained models appear to lack: (1) being able to decide when to talk; (2) producing coherent utterances grounded on multiple characters. We compare models trained on our new dataset to existing pairwise-trained dialogue models, as well as large language models with few-shot prompting. We find that our new dataset, MultiLIGHT, which we will publicly release, can help bring significant improvements in the group setting.
♻ ☆ Introduction to Medical Imaging Informatics
Medical imaging informatics is a rapidly growing field that combines the principles of medical imaging and informatics to improve the acquisition, management, and interpretation of medical images. This chapter introduces the basic concepts of medical imaging informatics, including image processing, feature engineering, and machine learning. It also discusses the recent advancements in computer vision and deep learning technologies and how they are used to develop new quantitative image markers and prediction models for disease detection, diagnosis, and prognosis prediction. By covering the basic knowledge of medical imaging informatics, this chapter provides a foundation for understanding the role of informatics in medicine and its potential impact on patient care.
comment: 18 pages, 11 figures, 2 tables; Acceptance of the chapter for the Springer book "Data-driven approaches to medical imaging"
♻ ☆ Patient Dropout Prediction in Virtual Health: A Multimodal Dynamic Knowledge Graph and Text Mining Approach
Virtual health has been acclaimed as a transformative force in healthcare delivery. Yet, its dropout issue is critical that leads to poor health outcomes, increased health, societal, and economic costs. Timely prediction of patient dropout enables stakeholders to take proactive steps to address patients' concerns, potentially improving retention rates. In virtual health, the information asymmetries inherent in its delivery format, between different stakeholders, and across different healthcare delivery systems hinder the performance of existing predictive methods. To resolve those information asymmetries, we propose a Multimodal Dynamic Knowledge-driven Dropout Prediction (MDKDP) framework that learns implicit and explicit knowledge from doctor-patient dialogues and the dynamic and complex networks of various stakeholders in both online and offline healthcare delivery systems. We evaluate MDKDP by partnering with one of the largest virtual health platforms in China. MDKDP improves the F1-score by 3.26 percentage points relative to the best benchmark. Comprehensive robustness analyses show that integrating stakeholder attributes, knowledge dynamics, and compact bilinear pooling significantly improves the performance. Our work provides significant implications for healthcare IT by revealing the value of mining relations and knowledge across different service modalities. Practically, MDKDP offers a novel design artifact for virtual health platforms in patient dropout management.
♻ ☆ PyTorch Hyperparameter Tuning - A Tutorial for spotPython
The goal of hyperparameter tuning (or hyperparameter optimization) is to optimize the hyperparameters to improve the performance of the machine or deep learning model. spotPython (``Sequential Parameter Optimization Toolbox in Python'') is the Python version of the well-known hyperparameter tuner SPOT, which has been developed in the R programming environment for statistical analysis for over a decade. PyTorch is an optimized tensor library for deep learning using GPUs and CPUs. This document shows how to integrate the spotPython hyperparameter tuner into the PyTorch training workflow. As an example, the results of the CIFAR10 image classifier are used. In addition to an introduction to spotPython, this tutorial also includes a brief comparison with Ray Tune, a Python library for running experiments and tuning hyperparameters. This comparison is based on the PyTorch hyperparameter tuning tutorial. The advantages and disadvantages of both approaches are discussed. We show that spotPython achieves similar or even better results while being more flexible and transparent than Ray Tune.
comment: Refers to spotPython version 0.2.15
♻ ☆ Convergence Analysis of Sequencial Split Learning on Heterogeneous Data
Federated Learning (FL) and Split Learning (SL) are two popular paradigms of distributed machine learning. By offloading the computation-intensive portions to the server, SL is promising for deep model training on resource-constrained devices, yet still lacking of rigorous convergence analysis. In this paper, we derive the convergence guarantees of Sequential SL (SSL, the vanilla case of SL that conducts the model training in sequence) for strongly/general/non-convex objectives on heterogeneous data. Notably, the derived guarantees suggest that SSL is better than Federated Averaging (FedAvg, the most popular algorithm in FL) on heterogeneous data. We validate the counterintuitive analysis result empirically on extremely heterogeneous data.
♻ ☆ HeterPS: Distributed Deep Learning With Reinforcement Learning Based Scheduling in Heterogeneous Environments
Deep neural networks (DNNs) exploit many layers and a large number of parameters to achieve excellent performance. The training process of DNN models generally handles large-scale input data with many sparse features, which incurs high Input/Output (IO) cost, while some layers are compute-intensive. The training process generally exploits distributed computing resources to reduce training time. In addition, heterogeneous computing resources, e.g., CPUs, GPUs of multiple types, are available for the distributed training process. Thus, the scheduling of multiple layers to diverse computing resources is critical for the training process. To efficiently train a DNN model using the heterogeneous computing resources, we propose a distributed framework, i.e., Paddle-Heterogeneous Parameter Server (Paddle-HeterPS), composed of a distributed architecture and a Reinforcement Learning (RL)-based scheduling method. The advantages of Paddle-HeterPS are three-fold compared with existing frameworks. First, Paddle-HeterPS enables efficient training process of diverse workloads with heterogeneous computing resources. Second, Paddle-HeterPS exploits an RL-based method to efficiently schedule the workload of each layer to appropriate computing resources to minimize the cost while satisfying throughput constraints. Third, Paddle-HeterPS manages data storage and data communication among distributed computing resources. We carry out extensive experiments to show that Paddle-HeterPS significantly outperforms state-of-the-art approaches in terms of throughput (14.5 times higher) and monetary cost (312.3% smaller). The codes of the framework are publicly available at: https://github.com/PaddlePaddle/Paddle.
comment: 14 pages, 11 figures, 2 tables; To appear in Future Generation Computer Systems (FGCS)
♻ ☆ Adversarially Robust PAC Learnability of Real-Valued Functions ICML2023
We study robustness to test-time adversarial attacks in the regression setting with $\ell_p$ losses and arbitrary perturbation sets. We address the question of which function classes are PAC learnable in this setting. We show that classes of finite fat-shattering dimension are learnable in both realizable and agnostic settings. Moreover, for convex function classes, they are even properly learnable. In contrast, some non-convex function classes provably require improper learning algorithms. Our main technique is based on a construction of an adversarially robust sample compression scheme of a size determined by the fat-shattering dimension. Along the way, we introduce a novel agnostic sample compression scheme for real-valued functions, which may be of independent interest.
comment: accepted to ICML2023
♻ ☆ Tracr: Compiled Transformers as a Laboratory for Interpretability
We show how to "compile" human-readable programs into standard decoder-only transformer models. Our compiler, Tracr, generates models with known structure. This structure can be used to design experiments. For example, we use it to study "superposition" in transformers that execute multi-step algorithms. Additionally, the known structure of Tracr-compiled models can serve as ground-truth for evaluating interpretability methods. Commonly, because the "programs" learned by transformers are unknown it is unclear whether an interpretation succeeded. We demonstrate our approach by implementing and examining programs including computing token frequencies, sorting, and parenthesis checking. We provide an open-source implementation of Tracr at https://github.com/deepmind/tracr.
♻ ☆ GAN-MPC: Training Model Predictive Controllers with Parameterized Cost Functions using Demonstrations from Non-identical Experts
Model predictive control (MPC) is a popular approach for trajectory optimization in practical robotics applications. MPC policies can optimize trajectory parameters under kinodynamic and safety constraints and provide guarantees on safety, optimality, generalizability, interpretability, and explainability. However, some behaviors are complex and it is difficult to hand-craft an MPC objective function. A special class of MPC policies called Learnable-MPC addresses this difficulty using imitation learning from expert demonstrations. However, they require the demonstrator and the imitator agents to be identical which is hard to satisfy in many real world applications of robotics. In this paper, we address the practical problem of training Learnable-MPC policies when the demonstrator and the imitator do not share the same dynamics and their state spaces may have a partial overlap. We propose a novel approach that uses a generative adversarial network (GAN) to minimize the Jensen-Shannon divergence between the state-trajectory distributions of the demonstrator and the imitator. We evaluate our approach on a variety of simulated robotics tasks of DeepMind Control suite and demonstrate the efficacy of our approach at learning the demonstrator's behavior without having to copy their actions.
comment: Recipient of the best paper award at RBCDSAI-DAI 2023, IIT Madras (https://rbcdsai.iitm.ac.in/DAI-2023/)
♻ ☆ On the Role of Randomization in Adversarially Robust Classification
Deep neural networks are known to be vulnerable to small adversarial perturbations in test data. To defend against adversarial attacks, probabilistic classifiers have been proposed as an alternative to deterministic ones. However, literature has conflicting findings on the effectiveness of probabilistic classifiers in comparison to deterministic ones. In this paper, we clarify the role of randomization in building adversarially robust classifiers. Given a base hypothesis set of deterministic classifiers, we show the conditions under which a randomized ensemble outperforms the hypothesis set in adversarial risk, extending previous results. Additionally, we show that for any probabilistic classifier (including randomized ensembles), there exists a deterministic classifier that outperforms it. Finally, we give an explicit description of the deterministic hypothesis set that contains such a deterministic classifier for many types of commonly used probabilistic classifiers, i.e. randomized ensembles and parametric/input noise injection.
comment: 9 pages + bibliography and appendix, 2 figures. This is a replacement with important changes, including a refinement of the main result in the last paper and a new section on passing from deterministic to randomized
♻ ☆ Improving Expressivity of GNNs with Subgraph-specific Factor Embedded Normalization
Graph Neural Networks~(GNNs) have emerged as a powerful category of learning architecture for handling graph-structured data. However, existing GNNs typically ignore crucial structural characteristics in node-induced subgraphs, which thus limits their expressiveness for various downstream tasks. In this paper, we strive to strengthen the representative capabilities of GNNs by devising a dedicated plug-and-play normalization scheme, termed as SUbgraph-sPEcific FactoR Embedded Normalization (SuperNorm), that explicitly considers the intra-connection information within each node-induced subgraph. To this end, we embed the subgraph-specific factor at the beginning and the end of the standard BatchNorm, as well as incorporate graph instance-specific statistics for improved distinguishable capabilities. In the meantime, we provide theoretical analysis to support that, with the elaborated SuperNorm, an arbitrary GNN is at least as powerful as the 1-WL test in distinguishing non-isomorphism graphs. Furthermore, the proposed SuperNorm scheme is also demonstrated to alleviate the over-smoothing phenomenon. Experimental results related to predictions of graph, node, and link properties on the eight popular datasets demonstrate the effectiveness of the proposed method. The code is available at \url{https://github.com/chenchkx/SuperNorm}.
comment: 13 pages, 7 figures
♻ ☆ Learning Gaussian Mixture Representations for Tensor Time Series Forecasting IJCAI 2023
Tensor time series (TTS) data, a generalization of one-dimensional time series on a high-dimensional space, is ubiquitous in real-world scenarios, especially in monitoring systems involving multi-source spatio-temporal data (e.g., transportation demands and air pollutants). Compared to modeling time series or multivariate time series, which has received much attention and achieved tremendous progress in recent years, tensor time series has been paid less effort. Properly coping with the tensor time series is a much more challenging task, due to its high-dimensional and complex inner structure. In this paper, we develop a novel TTS forecasting framework, which seeks to individually model each heterogeneity component implied in the time, the location, and the source variables. We name this framework as GMRL, short for Gaussian Mixture Representation Learning. Experiment results on two real-world TTS datasets verify the superiority of our approach compared with the state-of-the-art baselines. Code and data are published on https://github.com/beginner-sketch/GMRL.
comment: Accepted by IJCAI 2023 Main Track
♻ ☆ Group Fairness with Uncertainty in Sensitive Attributes
Learning a fair predictive model is crucial to mitigate biased decisions against minority groups in high-stakes applications. A common approach to learn such a model involves solving an optimization problem that maximizes the predictive power of the model under an appropriate group fairness constraint. However, in practice, sensitive attributes are often missing or noisy resulting in uncertainty. We demonstrate that solely enforcing fairness constraints on uncertain sensitive attributes can fall significantly short in achieving the level of fairness of models trained without uncertainty. To overcome this limitation, we propose a bootstrap-based algorithm that achieves the target level of fairness despite the uncertainty in sensitive attributes. The algorithm is guided by a Gaussian analysis for the independence notion of fairness where we propose a robust quadratically constrained quadratic problem to ensure a strict fairness guarantee with uncertain sensitive attributes. Our algorithm is applicable to both discrete and continuous sensitive attributes and is effective in real-world classification and regression tasks for various group fairness notions, e.g., independence and separation.
♻ ☆ Open-TransMind: A New Baseline and Benchmark for 1st Foundation Model Challenge of Intelligent Transportation
With the continuous improvement of computing power and deep learning algorithms in recent years, the foundation model has grown in popularity. Because of its powerful capabilities and excellent performance, this technology is being adopted and applied by an increasing number of industries. In the intelligent transportation industry, artificial intelligence faces the following typical challenges: few shots, poor generalization, and a lack of multi-modal techniques. Foundation model technology can significantly alleviate the aforementioned issues. To address these, we designed the 1st Foundation Model Challenge, with the goal of increasing the popularity of foundation model technology in traffic scenarios and promoting the rapid development of the intelligent transportation industry. The challenge is divided into two tracks: all-in-one and cross-modal image retrieval. Furthermore, we provide a new baseline and benchmark for the two tracks, called Open-TransMind. According to our knowledge, Open-TransMind is the first open-source transportation foundation model with multi-task and multi-modal capabilities. Simultaneously, Open-TransMind can achieve state-of-the-art performance on detection, classification, and segmentation datasets of traffic scenarios. Our source code is available at https://github.com/Traffic-X/Open-TransMind.
♻ ☆ Policy Gradient in Robust MDPs with Global Convergence Guarantee
Robust Markov decision processes (RMDPs) provide a promising framework for computing reliable policies in the face of model errors. Many successful reinforcement learning algorithms build on variations of policy-gradient methods, but adapting these methods to RMDPs has been challenging. As a result, the applicability of RMDPs to large, practical domains remains limited. This paper proposes a new Double-Loop Robust Policy Gradient (DRPG), the first generic policy gradient method for RMDPs. In contrast with prior robust policy gradient algorithms, DRPG monotonically reduces approximation errors to guarantee convergence to a globally optimal policy in tabular RMDPs. We introduce a novel parametric transition kernel and solve the inner loop robust policy via a gradient-based method. Finally, our numerical results demonstrate the utility of our new algorithm and confirm its global convergence properties.
♻ ☆ Temporal Difference Learning with Continuous Time and State in the Stochastic Setting
We consider the problem of continuous-time policy evaluation. This consists in learning through observations the value function associated with an uncontrolled continuous-time stochastic dynamic and a reward function. We propose two original variants of the well-known TD(0) method using vanishing time steps. One is model-free and the other is model-based. For both methods, we prove theoretical convergence rates that we subsequently verify through numerical simulations. Alternatively, those methods can be interpreted as novel reinforcement learning approaches for approximating solutions of linear PDEs (partial differential equations) or linear BSDEs (backward stochastic differential equations).
♻ ☆ Two Losses Are Better Than One: Faster Optimization Using a Cheaper Proxy
We present an algorithm for minimizing an objective with hard-to-compute gradients by using a related, easier-to-access function as a proxy. Our algorithm is based on approximate proximal point iterations on the proxy combined with relatively few stochastic gradients from the objective. When the difference between the objective and the proxy is $\delta$-smooth, our algorithm guarantees convergence at a rate matching stochastic gradient descent on a $\delta$-smooth objective, which can lead to substantially better sample efficiency. Our algorithm has many potential applications in machine learning, and provides a principled means of leveraging synthetic data, physics simulators, mixed public and private data, and more.
♻ ☆ Vid2Act: Activate Offline Videos for Visual RL
Pretraining RL models on offline video datasets is a promising way to improve their training efficiency in online tasks, but challenging due to the inherent mismatch in tasks, dynamics, and behaviors across domains. A recent model, APV, sidesteps the accompanied action records in offline datasets and instead focuses on pretraining a task-irrelevant, action-free world model within the source domains. We present Vid2Act, a model-based RL method that learns to transfer valuable action-conditioned dynamics and potentially useful action demonstrations from offline to online settings. The main idea is to use the world models not only as simulators for behavior learning but also as tools to measure the domain relevance for both dynamics representation transfer and policy transfer. Specifically, we train the world models to generate a set of time-varying task similarities using a domain-selective knowledge distillation loss. These similarities serve two purposes: (i) adaptively transferring the most useful source knowledge to facilitate dynamics learning, and (ii) learning to replay the most relevant source actions to guide the target policy. We demonstrate the advantages of Vid2Act over the action-free visual RL pretraining method in both Meta-World and DeepMind Control Suite.
♻ ☆ ROIPCA: An online memory-restricted PCA algorithm based on rank-one updates
Principal components analysis (PCA) is a fundamental algorithm in data analysis. Its memory-restricted online versions are useful in many modern applications, where the data are too large to fit in memory, or when data arrive as a stream of items. In this paper, we propose ROIPCA and fROIPCA, two online PCA algorithms that are based on rank-one updates. While ROIPCA is typically more accurate, fROIPCA is faster and has comparable accuracy. We show the relation between fROIPCA and an existing popular gradient algorithm for online PCA, and in particular, prove that fROIPCA is in fact a gradient algorithm with an optimal learning rate. We demonstrate numerically the advantages of our algorithms over existing state-of-the-art algorithms in terms of accuracy and runtime.
comment: 23 pages, 2 figures
♻ ☆ Planning Multiple Epidemic Interventions with Reinforcement Learning
Combating an epidemic entails finding a plan that describes when and how to apply different interventions, such as mask-wearing mandates, vaccinations, school or workplace closures. An optimal plan will curb an epidemic with minimal loss of life, disease burden, and economic cost. Finding an optimal plan is an intractable computational problem in realistic settings. Policy-makers, however, would greatly benefit from tools that can efficiently search for plans that minimize disease and economic costs especially when considering multiple possible interventions over a continuous and complex action space given a continuous and equally complex state space. We formulate this problem as a Markov decision process. Our formulation is unique in its ability to represent multiple continuous interventions over any disease model defined by ordinary differential equations. We illustrate how to effectively apply state-of-the-art actor-critic reinforcement learning algorithms (PPO and SAC) to search for plans that minimize overall costs. We empirically evaluate the learning performance of these algorithms and compare their performance to hand-crafted baselines that mimic plans constructed by policy-makers. Our method outperforms baselines. Our work confirms the viability of a computational approach to support policy-makers
♻ ☆ A Large-Scale Study of Probabilistic Calibration in Neural Network Regression ICML 2023
Accurate probabilistic predictions are essential for optimal decision making. While neural network miscalibration has been studied primarily in classification, we investigate this in the less-explored domain of regression. We conduct the largest empirical study to date to assess the probabilistic calibration of neural networks. We also analyze the performance of recalibration, conformal, and regularization methods to enhance probabilistic calibration. Additionally, we introduce novel differentiable recalibration and regularization methods, uncovering new insights into their effectiveness. Our findings reveal that regularization methods offer a favorable tradeoff between calibration and sharpness. Post-hoc methods exhibit superior probabilistic calibration, which we attribute to the finite-sample coverage guarantee of conformal prediction. Furthermore, we demonstrate that quantile recalibration can be considered as a specific case of conformal prediction. Our study is fully reproducible and implemented in a common code base for fair comparisons.
comment: Accepted at the 40th International Conference on Machine Learning (ICML 2023)
♻ ☆ Digital Audio Forensics: Blind Human Voice Mimicry Detection
Audio is one of the most used ways of human communication, but at the same time it can be easily misused to trick people. With the revolution of AI, the related technologies are now accessible to almost everyone thus making it simple for the criminals to commit crimes and forgeries. In this work, we introduce a deep learning method to develop a classifier that will blindly classify an input audio as real or mimicked; the word 'blindly' refers to the ability to detect mimicked audio without references or real sources. The proposed model was trained on a set of important features extracted from a large dataset of audios to get a classifier that was tested on the same set of features from different audios. The data was extracted from two raw datasets, especially composed for this work; an all English dataset and a mixed dataset (Arabic plus English). These datasets have been made available, in raw form, through GitHub for the use of the research community at https://github.com/SaSs7/Dataset. For the purpose of comparison, the audios were also classified through human inspection with the subjects being the native speakers. The ensued results were interesting and exhibited formidable accuracy.
comment: 14 pages, 4 figures (6 if you count subfigures), 2 tables
♻ ☆ Boosting Tail Neural Network for Realtime Custom Keyword Spotting
In this paper, we propose a Boosting Tail Neural Network (BTNN) for improving the performance of Realtime Custom Keyword Spotting (RCKS) that is still an industrial challenge for demanding powerful classification ability with limited computation resources. Inspired by Brain Science that a brain is only partly activated for a nerve simulation and numerous machine learning algorithms are developed to use a batch of weak classifiers to resolve arduous problems, which are often proved to be effective. We show that this method is helpful to the RCKS problem. The proposed approach achieve better performances in terms of wakeup rate and false alarm. In our experiments compared with those traditional algorithms that use only one strong classifier, it gets 18\% relative improvement. We also point out that this approach may be promising in future ASR exploration.
comment: 4 pages, 8 figures, 2 tables
♻ ☆ A Trustworthiness Score to Evaluate CNNs Predictions
Due to the black box nature of Convolutional Neural Networks (CNNs), the continuous validation of CNNs during operation is challenging with the absence of a human monitor. As a result this makes it difficult for developers and regulators to gain confidence in the deployment of autonomous systems employing CNNs. It is critical for safety during operation to know when CNN's predictions are trustworthy or suspicious. With the absence of a human monitor, the basic approach is to use the model's output confidence score to assess if predictions are trustworthy or suspicious. However, the model's confidence score is a result of computations coming from a black box, therefore lacks transparency and makes it challenging to automatedly credit trustworthiness to predictions. We introduce the trustworthiness score (TS), a simple metric that provides a more transparent and effective way of providing confidence in CNNs predictions compared to model's confidence score. The metric quantifies the trustworthiness in a prediction by checking for the existence of certain features in the predictions made by the CNN. We also use the underlying idea of the TS metric, to provide a suspiciousness score (SS) in the overall input frame to help in the detection of suspicious frames where false negatives exist. We conduct a case study using YOLOv5 on persons detection to demonstrate our method and usage of TS and SS. The case study shows that using our method consistently improves the precision of predictions compared to relying on model confidence score alone, for both 1) approving of trustworthy predictions (~20% improvement) and 2) detecting suspicious frames (~5% improvement).
♻ ☆ Literature Review: Computer Vision Applications in Transportation Logistics and Warehousing
Computer vision applications in transportation logistics and warehousing have a huge potential for process automation. We present a structured literature review on research in the field to help leverage this potential. The literature is categorized w.r.t. the application, i.e. the task it tackles and w.r.t. the computer vision techniques that are used. Regarding applications, we subdivide the literature in two areas: Monitoring, i.e. observing and retrieving relevant information from the environment, and manipulation, where approaches are used to analyze and interact with the environment. Additionally, we point out directions for future research and link to recent developments in computer vision that are suitable for application in logistics. Finally, we present an overview of existing datasets and industrial solutions. The results of our analysis are also available online at https://a-nau.github.io/cv-in-logistics.
♻ ☆ SGD with Large Step Sizes Learns Sparse Features ICML 2023
We showcase important features of the dynamics of the Stochastic Gradient Descent (SGD) in the training of neural networks. We present empirical observations that commonly used large step sizes (i) lead the iterates to jump from one side of a valley to the other causing loss stabilization, and (ii) this stabilization induces a hidden stochastic dynamics orthogonal to the bouncing directions that biases it implicitly toward sparse predictors. Furthermore, we show empirically that the longer large step sizes keep SGD high in the loss landscape valleys, the better the implicit regularization can operate and find sparse representations. Notably, no explicit regularization is used so that the regularization effect comes solely from the SGD training dynamics influenced by the step size schedule. Therefore, these observations unveil how, through the step size schedules, both gradient and noise drive together the SGD dynamics through the loss landscape of neural networks. We justify these findings theoretically through the study of simple neural network models as well as qualitative arguments inspired from stochastic processes. Finally, this analysis allows us to shed a new light on some common practice and observed phenomena when training neural networks. The code of our experiments is available at https://github.com/tml-epfl/sgd-sparse-features.
comment: The camera-ready version (ICML 2023): extended experiments on deep networks (DenseNets on CIFAR-10, CIFAR-100, and Tiny ImageNet), empirically validated the SDE modelling, improved the clarity of the paper
♻ ☆ A Deep Learning Framework for Verilog Autocompletion Towards Design and Verification Automation
Innovative Electronic Design Automation (EDA) solutions are important to meet the design requirements for increasingly complex electronic devices. Verilog, a hardware description language, is widely used for the design and verification of digital circuits and is synthesized using specific EDA tools. However, writing code is a repetitive and time-intensive task. This paper proposes, primarily, a novel deep learning framework for training a Verilog autocompletion model and, secondarily, a Verilog dataset of files and snippets obtained from open-source repositories. The framework involves integrating models pretrained on general programming language data and finetuning them on a dataset curated to be similar to a target downstream task. This is validated by comparing different pretrained models trained on different subsets of the proposed Verilog dataset using multiple evaluation metrics. These experiments demonstrate that the proposed framework achieves better BLEU, ROUGE-L, and chrF scores by 9.5%, 6.7%, and 6.9%, respectively, compared to a model trained from scratch. Code and data are made available at: https://github.com/99EnriqueD/verilog_autocompletion .
comment: Updated text to correct language errors and added a link to supplementary code and data (https://github.com/99EnriqueD/verilog_autocompletion). 6 pages, 3 figures, 4 tables. To be presented as a WIP poster at DAC 2023
♻ ☆ Machine Learning Testing in an ADAS Case Study Using Simulation-Integrated Bio-Inspired Search-Based Testing
This paper presents an extended version of Deeper, a search-based simulation-integrated test solution that generates failure-revealing test scenarios for testing a deep neural network-based lane-keeping system. In the newly proposed version, we utilize a new set of bio-inspired search algorithms, genetic algorithm (GA), $({\mu}+{\lambda})$ and $({\mu},{\lambda})$ evolution strategies (ES), and particle swarm optimization (PSO), that leverage a quality population seed and domain-specific cross-over and mutation operations tailored for the presentation model used for modeling the test scenarios. In order to demonstrate the capabilities of the new test generators within Deeper, we carry out an empirical evaluation and comparison with regard to the results of five participating tools in the cyber-physical systems testing competition at SBST 2021. Our evaluation shows the newly proposed test generators in Deeper not only represent a considerable improvement on the previous version but also prove to be effective and efficient in provoking a considerable number of diverse failure-revealing test scenarios for testing an ML-driven lane-keeping system. They can trigger several failures while promoting test scenario diversity, under a limited test time budget, high target failure severity, and strict speed limit constraints.
comment: Accepted for publication in Journal Of Software: Evolution And Process
♻ ☆ Differentially Private Distributed Bayesian Linear Regression with MCMC
We propose a novel Bayesian inference framework for distributed differentially private linear regression. We consider a distributed setting where multiple parties hold parts of the data and share certain summary statistics of their portions in privacy-preserving noise. We develop a novel generative statistical model for privately shared statistics, which exploits a useful distributional relation between the summary statistics of linear regression. Bayesian estimation of the regression coefficients is conducted mainly using Markov chain Monte Carlo algorithms, while we also provide a fast version to perform Bayesian estimation in one iteration. The proposed methods have computational advantages over their competitors. We provide numerical results on both real and simulated data, which demonstrate that the proposed algorithms provide well-rounded estimation and prediction.
comment: 15 pages, 4 figures, code available at: https://github.com/sinanyildirim/Bayesian_DP_dist_LR
♻ ☆ A Modern Look at the Relationship between Sharpness and Generalization ICML 2023
Sharpness of minima is a promising quantity that can correlate with generalization in deep networks and, when optimized during training, can improve generalization. However, standard sharpness is not invariant under reparametrizations of neural networks, and, to fix this, reparametrization-invariant sharpness definitions have been proposed, most prominently adaptive sharpness (Kwon et al., 2021). But does it really capture generalization in modern practical settings? We comprehensively explore this question in a detailed study of various definitions of adaptive sharpness in settings ranging from training from scratch on ImageNet and CIFAR-10 to fine-tuning CLIP on ImageNet and BERT on MNLI. We focus mostly on transformers for which little is known in terms of sharpness despite their widespread usage. Overall, we observe that sharpness does not correlate well with generalization but rather with some training parameters like the learning rate that can be positively or negatively correlated with generalization depending on the setup. Interestingly, in multiple cases, we observe a consistent negative correlation of sharpness with out-of-distribution error implying that sharper minima can generalize better. Finally, we illustrate on a simple model that the right sharpness measure is highly data-dependent, and that we do not understand well this aspect for realistic data distributions. The code of our experiments is available at https://github.com/tml-epfl/sharpness-vs-generalization.
comment: The camera-ready version (accepted at ICML 2023)
♻ ☆ Rigid Body Flows for Sampling Molecular Crystal Structures
Normalizing flows (NF) are a class of powerful generative models that have gained popularity in recent years due to their ability to model complex distributions with high flexibility and expressiveness. In this work, we introduce a new type of normalizing flow that is tailored for modeling positions and orientations of multiple objects in three-dimensional space, such as molecules in a crystal. Our approach is based on two key ideas: first, we define smooth and expressive flows on the group of unit quaternions, which allows us to capture the continuous rotational motion of rigid bodies; second, we use the double cover property of unit quaternions to define a proper density on the rotation group. This ensures that our model can be trained using standard likelihood-based methods or variational inference with respect to a thermodynamic target density. We evaluate the method by training Boltzmann generators for two molecular examples, namely the multi-modal density of a tetrahedral system in an external field and the ice XI phase in the TIP4P water model. Our flows can be combined with flows operating on the internal degrees of freedom of molecules and constitute an important step towards the modeling of distributions of many interacting molecules.
comment: International Conference on Machine Learning, 2023
♻ ☆ Gaussian Hierarchical Latent Dirichlet Allocation: Bringing Polysemy Back
Topic models are widely used to discover the latent representation of a set of documents. The two canonical models are latent Dirichlet allocation, and Gaussian latent Dirichlet allocation, where the former uses multinomial distributions over words, and the latter uses multivariate Gaussian distributions over pre-trained word embedding vectors as the latent topic representations, respectively. Compared with latent Dirichlet allocation, Gaussian latent Dirichlet allocation is limited in the sense that it does not capture the polysemy of a word such as ``bank.'' In this paper, we show that Gaussian latent Dirichlet allocation could recover the ability to capture polysemy by introducing a hierarchical structure in the set of topics that the model can use to represent a given document. Our Gaussian hierarchical latent Dirichlet allocation significantly improves polysemy detection compared with Gaussian-based models and provides more parsimonious topic representations compared with hierarchical latent Dirichlet allocation. Our extensive quantitative experiments show that our model also achieves better topic coherence and held-out document predictive accuracy over a wide range of corpus and word embedding vectors.
♻ ☆ Random Grid Neural Processes for Parametric Partial Differential Equations
We introduce a new class of spatially stochastic physics and data informed deep latent models for parametric partial differential equations (PDEs) which operate through scalable variational neural processes. We achieve this by assigning probability measures to the spatial domain, which allows us to treat collocation grids probabilistically as random variables to be marginalised out. Adapting this spatial statistics view, we solve forward and inverse problems for parametric PDEs in a way that leads to the construction of Gaussian process models of solution fields. The implementation of these random grids poses a unique set of challenges for inverse physics informed deep learning frameworks and we propose a new architecture called Grid Invariant Convolutional Networks (GICNets) to overcome these challenges. We further show how to incorporate noisy data in a principled manner into our physics informed model to improve predictions for problems where data may be available but whose measurement location does not coincide with any fixed mesh or grid. The proposed method is tested on a nonlinear Poisson problem, Burgers equation, and Navier-Stokes equations, and we provide extensive numerical comparisons. We demonstrate significant computational advantages over current physics informed neural learning methods for parametric PDEs while improving the predictive capabilities and flexibility of these models.
♻ ☆ Dual Propagation: Accelerating Contrastive Hebbian Learning with Dyadic Neurons
Activity difference based learning algorithms-such as contrastive Hebbian learning and equilibrium propagation-have been proposed as biologically plausible alternatives to error back-propagation. However, on traditional digital chips these algorithms suffer from having to solve a costly inference problem twice, making these approaches more than two orders of magnitude slower than back-propagation. In the analog realm equilibrium propagation may be promising for fast and energy efficient learning, but states still need to be inferred and stored twice. Inspired by lifted neural networks and compartmental neuron models we propose a simple energy based compartmental neuron model, termed dual propagation, in which each neuron is a dyad with two intrinsic states. At inference time these intrinsic states encode the error/activity duality through their difference and their mean respectively. The advantage of this method is that only a single inference phase is needed and that inference can be solved in layerwise closed-form. Experimentally we show on common computer vision datasets, including Imagenet32x32, that dual propagation performs equivalently to back-propagation both in terms of accuracy and runtime.
comment: Added reflections on biological plausibility and results comparisons to state-of-the-art versions of equilibrium propagation and difference target propagation
♻ ☆ Federated Deep Learning for Intrusion Detection in IoT Networks
The vast increase of IoT technologies and the ever-evolving attack vectors and threat actors have increased cyber-security risks dramatically. Novel attacks can compromise IoT devices to gain access to sensitive data or control them to deploy further malicious activities. The detection of novel attacks often relies upon AI solutions. A common approach to implementing AI-based IDS in distributed IoT systems is in a centralised manner. However, this approach may violate data privacy and secrecy. In addition, centralised data collection prohibits the scale-up of IDSs. Therefore, intrusion detection solutions in IoT ecosystems need to move towards a decentralised direction. FL has attracted significant interest in recent years due to its ability to perform collaborative learning while preserving data confidentiality and locality. Nevertheless, most FL-based IDS for IoT systems are designed under unrealistic data distribution conditions. To that end, we design an experiment representative of the real world and evaluate the performance of two FL IDS implementations, one based on DNNs and another on our previous work on DBNs. For our experiments, we rely on TON-IoT, a realistic IoT network traffic dataset, associating each IP address with a single FL client. Additionally, we explore pre-training and investigate various aggregation methods to mitigate the impact of data heterogeneity. Lastly, we benchmark our approach against a centralised solution. The comparison shows that the heterogeneous nature of the data has a considerable negative impact on the model performance when trained in a distributed manner. However, in the case of a pre-trained initial global FL model, we demonstrate a performance improvement of over 20% (F1-score) when compared against a randomly initiated global model.
comment: 14 pages, 5 figues, 3 tables
♻ ☆ Neural Collapse in Deep Linear Networks: From Balanced to Imbalanced Data
Modern deep neural networks have achieved impressive performance on tasks from image classification to natural language processing. Surprisingly, these complex systems with massive amounts of parameters exhibit the same structural properties in their last-layer features and classifiers across canonical datasets when training until convergence. In particular, it has been observed that the last-layer features collapse to their class-means, and those class-means are the vertices of a simplex Equiangular Tight Frame (ETF). This phenomenon is known as Neural Collapse ($\mathcal{NC}$). Recent papers have theoretically shown that $\mathcal{NC}$ emerges in the global minimizers of training problems with the simplified ``unconstrained feature model''. In this context, we take a step further and prove the $\mathcal{NC}$ occurrences in deep linear networks for the popular mean squared error (MSE) and cross entropy (CE) losses, showing that global solutions exhibit $\mathcal{NC}$ properties across the linear layers. Furthermore, we extend our study to imbalanced data for MSE loss and present the first geometric analysis of $\mathcal{NC}$ under bias-free setting. Our results demonstrate the convergence of the last-layer features and classifiers to a geometry consisting of orthogonal vectors, whose lengths depend on the amount of data in their corresponding classes. Finally, we empirically validate our theoretical analyses on synthetic and practical network architectures with both balanced and imbalanced scenarios.
comment: 75 pages, 20 figures, 4 tables. Hien Dang and Tho Tran contributed equally to this work
♻ ☆ Quantum Machine Learning for Malware Classification
In a context of malicious software detection, machine learning (ML) is widely used to generalize to new malware. However, it has been demonstrated that ML models can be fooled or may have generalization problems on malware that has never been seen. We investigate the possible benefits of quantum algorithms for classification tasks. We implement two models of Quantum Machine Learning algorithms, and we compare them to classical models for the classification of a dataset composed of malicious and benign executable files. We try to optimize our algorithms based on methods found in the literature, and analyze our results in an exploratory way, to identify the most interesting directions to explore for the future.
♻ ☆ PCT-CycleGAN: Paired Complementary Temporal Cycle-Consistent Adversarial Networks for Radar-Based Precipitation Nowcasting
The precipitation nowcasting methods have been elaborated over the centuries because rain has a crucial impact on human life. Not only quantitative precipitation forecast (QPF) models and convolutional long short-term memory (ConvLSTM), but also various sophisticated methods such as the latest MetNet-2 are emerging. In this paper, we propose a paired complementary temporal cycle-consistent adversarial networks (PCT-CycleGAN) for radar-based precipitation nowcasting, inspired by cycle-consistent adversarial networks (CycleGAN), which shows strong performance in image-to-image translation. PCT-CycleGAN generates temporal causality using two generator networks with forward and backward temporal dynamics in paired complementary cycles. Each generator network learns a huge number of one-to-one mappings about time-dependent radar-based precipitation data to approximate a mapping function representing the temporal dynamics in each direction. To create robust temporal causality between paired complementary cycles, novel connection loss is proposed. And torrential loss to cover exceptional heavy rain events is also proposed. The generator network learning forward temporal dynamics in PCT-CycleGAN generates radar-based precipitation data 10 minutes from the current time. Also, it provides a reliable prediction of up to 2 hours with iterative forecasting. The superiority of PCT-CycleGAN is demonstrated through qualitative and quantitative comparisons with several previous methods.
♻ ☆ Explaining the Explainers in Graph Neural Networks: a Comparative Study
Following a fast initial breakthrough in graph based learning, Graph Neural Networks (GNNs) have reached a widespread application in many science and engineering fields, prompting the need for methods to understand their decision process. GNN explainers have started to emerge in recent years, with a multitude of methods both novel or adapted from other domains. To sort out this plethora of alternative approaches, several studies have benchmarked the performance of different explainers in terms of various explainability metrics. However, these earlier works make no attempts at providing insights into why different GNN architectures are more or less explainable, or which explainer should be preferred in a given setting. In this survey, we fill these gaps by devising a systematic experimental study, which tests ten explainers on eight representative architectures trained on six carefully designed graph and node classification datasets. With our results we provide key insights on the choice and applicability of GNN explainers, we isolate key components that make them usable and successful and provide recommendations on how to avoid common interpretation pitfalls. We conclude by highlighting open questions and directions of possible future research.
♻ ☆ DevFormer: A Symmetric Transformer for Context-Aware Device Placement ICML
In this paper, we present DevFormer, a novel transformer-based architecture for addressing the complex and computationally demanding problem of hardware design optimization. Despite the demonstrated efficacy of transformers in domains including natural language processing and computer vision, their use in hardware design has been limited by the scarcity of offline data. Our approach addresses this limitation by introducing strong inductive biases such as relative positional embeddings and action-permutation symmetricity that effectively capture the hardware context and enable efficient design optimization with limited offline data. We apply DevFoemer to the problem of decoupling capacitor placement and show that it outperforms state-of-the-art methods in both simulated and real hardware, leading to improved performances while reducing the number of components by more than $30\%$. Finally, we show that our approach achieves promising results in other offline contextual learning-based combinatorial optimization tasks.
comment: International Conference on Machine Learning (ICML) 2023. Extended version of NeurIPS 2022 Offline RL Workshop "Collaborative symmetricity exploitation for offline learning of hardware design solver"
♻ ☆ Answering Complex Logical Queries on Knowledge Graphs via Query Computation Tree Optimization ICML 2023
Answering complex logical queries on incomplete knowledge graphs is a challenging task, and has been widely studied. Embedding-based methods require training on complex queries, and cannot generalize well to out-of-distribution query structures. Recent work frames this task as an end-to-end optimization problem, and it only requires a pretrained link predictor. However, due to the exponentially large combinatorial search space, the optimal solution can only be approximated, limiting the final accuracy. In this work, we propose QTO (Query Computation Tree Optimization) that can efficiently find the exact optimal solution. QTO finds the optimal solution by a forward-backward propagation on the tree-like computation graph, i.e., query computation tree. In particular, QTO utilizes the independence encoded in the query computation tree to reduce the search space, where only local computations are involved during the optimization procedure. Experiments on 3 datasets show that QTO obtains state-of-the-art performance on complex query answering, outperforming previous best results by an average of 22%. Moreover, QTO can interpret the intermediate solutions for each of the one-hop atoms in the query with over 90% accuracy. The code of our paper is at https://github.com/bys0318/QTO.
comment: To appear in ICML 2023
♻ ☆ Flat Seeking Bayesian Neural Networks
Bayesian Neural Networks (BNNs) provide a probabilistic interpretation for deep learning models by imposing a prior distribution over model parameters and inferring a posterior distribution based on observed data. The model sampled from the posterior distribution can be used for providing ensemble predictions and quantifying prediction uncertainty. It is well-known that deep learning models with lower sharpness have better generalization ability. However, existing posterior inferences are not aware of sharpness/flatness in terms of formulation, possibly leading to high sharpness for the models sampled from them. In this paper, we develop theories, the Bayesian setting, and the variational inference approach for the sharpness-aware posterior. Specifically, the models sampled from our sharpness-aware posterior, and the optimal approximate posterior estimating this sharpness-aware posterior, have better flatness, hence possibly possessing higher generalization ability. We conduct experiments by leveraging the sharpness-aware posterior with state-of-the-art Bayesian Neural Networks, showing that the flat-seeking counterparts outperform their baselines in all metrics of interest.
comment: Under review
♻ ☆ Revisiting Weighted Strategy for Non-stationary Parametric Bandits AISTATS 2023
Non-stationary parametric bandits have attracted much attention recently. There are three principled ways to deal with non-stationarity, including sliding-window, weighted, and restart strategies. As many non-stationary environments exhibit gradual drifting patterns, the weighted strategy is commonly adopted in real-world applications. However, previous theoretical studies show that its analysis is more involved and the algorithms are either computationally less efficient or statistically suboptimal. This paper revisits the weighted strategy for non-stationary parametric bandits. In linear bandits (LB), we discover that this undesirable feature is due to an inadequate regret analysis, which results in an overly complex algorithm design. We propose a refined analysis framework, which simplifies the derivation and importantly produces a simpler weight-based algorithm that is as efficient as window/restart-based algorithms while retaining the same regret as previous studies. Furthermore, our new framework can be used to improve regret bounds of other parametric bandits, including Generalized Linear Bandits (GLB) and Self-Concordant Bandits (SCB). For example, we develop a simple weighted GLB algorithm with an $\widetilde{O}(k_\mu^{\frac{5}{4}} c_\mu^{-\frac{3}{4}} d^{\frac{3}{4}} P_T^{\frac{1}{4}}T^{\frac{3}{4}})$ regret, improving the $\widetilde{O}(k_\mu^{2} c_\mu^{-1}d^{\frac{9}{10}} P_T^{\frac{1}{5}}T^{\frac{4}{5}})$ bound in prior work, where $k_\mu$ and $c_\mu$ characterize the reward model's nonlinearity, $P_T$ measures the non-stationarity, $d$ and $T$ denote the dimension and time horizon.
comment: AISTATS 2023
♻ ☆ Estimating 3D Dental Structures using Simulated Panoramic Radiographs and Neural Ray Tracing
Panoramic radiography (Panoramic X-ray, PX) is a widely used imaging modality for dental examination. Since PX only provides 2D flattened views of the oral structure, its applicability is limited as compared to 3D Cone-beam computed tomography (CBCT). In this paper, we propose a framework to estimate CBCT-like 3D structures from real-world PX. Our framework tackles full 3D reconstruction for varying subjects (patients) where each reconstruction is based only on a single panoramic image. We create an intermediate representation called simulated PX (SimPX) from CBCT data which is based both on the Beer-Lambert law of X-ray rendering and rotational principles of PX imaging. SimPX aims at not only truthfully simulating PX, but also facilitates the reverting process back to 3D data. We propose a novel neural model based on ray tracing which exploits both global and local input features to convert SimPX to 3D output. At inference, a real PX image is translated to a SimPX-style image with semantic regularization, and the translated image is processed by generation/refinement modules to produce high-quality outputs. Experiments show that our method outperforms prior state-of-the-art in reconstruction tasks both quantitatively and qualitatively. Our method does not require any prior information such as the shape of dental arches, nor the matched PX-CBCT dataset for training, which is difficult to obtain in clinical practice.
comment: 16 pages, 11 figures
♻ ☆ Test-Time Training on Nearest Neighbors for Large Language Models
Many recent efforts aim to augment language models with relevant information retrieved from a database at test time. We avoid the need for prompt engineering by directly fine-tuning the model on data retrieved at test time using its standard training setup. For this purpose, we build a large-scale distributed nearest neighbor index based on text embeddings of the Pile dataset. Given a query to a language model, our system retrieves the neighbors of the query and fine-tunes the model on the text data corresponding to those neighbors. Surprisingly, retrieving and training on as few as 20 neighbors, each for only one gradient iteration, drastically improves performance across more than twenty language modeling tasks in the Pile benchmark. For example, test-time training significantly narrows the performance gap between a small GPT2 model and a GPTNeo model, more than ten times larger, that was specifically trained to convergence on the Pile. Sufficient index quality and size, however, are important. Our work establishes a valuable first baseline for implementing test-time training in the context of large language models, opening the door to numerous promising research avenues.
comment: Corrected Figure 8. Code repository here: https://github.com/socialfoundations/tttlm
♻ ☆ Functional Equivalence and Path Connectivity of Reducible Hyperbolic Tangent Networks
Understanding the learning process of artificial neural networks requires clarifying the structure of the parameter space within which learning takes place. A neural network parameter's functional equivalence class is the set of parameters implementing the same input--output function. For many architectures, almost all parameters have a simple and well-documented functional equivalence class. However, there is also a vanishing minority of reducible parameters, with richer functional equivalence classes caused by redundancies among the network's units. In this paper, we give an algorithmic characterisation of unit redundancies and reducible functional equivalence classes for a single-hidden-layer hyperbolic tangent architecture. We show that such functional equivalence classes are piecewise-linear path-connected sets, and that for parameters with a majority of redundant units, the sets have a diameter of at most 7 linear segments.
comment: 15 pages, 3 figures
♻ ☆ Solving NP-hard Min-max Routing Problems as Sequential Generation with Equity Context
Min-max routing problems aim to minimize the maximum tour length among agents as they collaboratively visit all cities, i.e., the completion time. These problems include impactful real-world applications but are known as NP-hard. Existing methods are facing challenges, particularly in large-scale problems that require the coordination of numerous agents to cover thousands of cities. This paper proposes a new deep-learning framework to solve large-scale min-max routing problems. We model the simultaneous decision-making of multiple agents as a sequential generation process, allowing the utilization of scalable deep-learning models for sequential decision-making. In the sequentially approximated problem, we propose a scalable contextual Transformer model, Equity-Transformer, which generates sequential actions considering an equitable workload among other agents. The effectiveness of Equity-Transformer is demonstrated through its superior performance in two representative min-max routing tasks: the min-max multiple traveling salesman problem (min-max mTSP) and the min-max multiple pick-up and delivery problem (min-max mPDP). Notably, our method achieves significant reductions of runtime, approximately 335 times, and cost values of about 53% compared to a competitive heuristic (LKH3) in the case of 100 vehicles with 1,000 cities of mTSP. We provide reproducible source code: https://github.com/kaist-silab/equity-transformer
comment: 18 pages, 7 figures
♻ ☆ Meta-SAGE: Scale Meta-Learning Scheduled Adaptation with Guided Exploration for Mitigating Scale Shift on Combinatorial Optimization ICML
This paper proposes Meta-SAGE, a novel approach for improving the scalability of deep reinforcement learning models for combinatorial optimization (CO) tasks. Our method adapts pre-trained models to larger-scale problems in test time by suggesting two components: a scale meta-learner (SML) and scheduled adaptation with guided exploration (SAGE). First, SML transforms the context embedding for subsequent adaptation of SAGE based on scale information. Then, SAGE adjusts the model parameters dedicated to the context embedding for a specific instance. SAGE introduces locality bias, which encourages selecting nearby locations to determine the next location. The locality bias gradually decays as the model is adapted to the target instance. Results show that Meta-SAGE outperforms previous adaptation methods and significantly improves scalability in representative CO tasks. Our source code is available at https://github.com/kaist-silab/meta-sage
comment: 18 pages, 9 figures, International Conference on Machine Learning (ICML) 2023
♻ ☆ Using Machine Teaching to Investigate Human Assumptions when Teaching Reinforcement Learners
Successful teaching requires an assumption of how the learner learns - how the learner uses experiences from the world to update their internal states. We investigate what expectations people have about a learner when they teach them in an online manner using rewards and punishment. We focus on a common reinforcement learning method, Q-learning, and examine what assumptions people have using a behavioral experiment. To do so, we first establish a normative standard, by formulating the problem as a machine teaching optimization problem. To solve the machine teaching optimization problem, we use a deep learning approximation method which simulates learners in the environment and learns to predict how feedback affects the learner's internal states. What do people assume about a learner's learning and discount rates when they teach them an idealized exploration-exploitation task? In a behavioral experiment, we find that people can teach the task to Q-learners in a relatively efficient and effective manner when the learner uses a small value for its discounting rate and a large value for its learning rate. However, they still are suboptimal. We also find that providing people with real-time updates of how possible feedback would affect the Q-learner's internal states weakly helps them teach. Our results reveal how people teach using evaluative feedback and provide guidance for how engineers should design machine agents in a manner that is intuitive for people.
comment: 21 pages, 4 figures
♻ ☆ What Makes Data Suitable for a Locally Connected Neural Network? A Necessary and Sufficient Condition Based on Quantum Entanglement
The question of what makes a data distribution suitable for deep learning is a fundamental open problem. Focusing on locally connected neural networks (a prevalent family of architectures that includes convolutional and recurrent neural networks as well as local self-attention models), we address this problem by adopting theoretical tools from quantum physics. Our main theoretical result states that a certain locally connected neural network is capable of accurate prediction over a data distribution if and only if the data distribution admits low quantum entanglement under certain canonical partitions of features. As a practical application of this result, we derive a preprocessing method for enhancing the suitability of a data distribution to locally connected neural networks. Experiments with widespread models over various datasets demonstrate our findings. We hope that our use of quantum entanglement will encourage further adoption of tools from physics for formally reasoning about the relation between deep learning and real-world data.
♻ ☆ Improved Privacy-Preserving PCA Using Space-optimized Homomorphic Matrix Multiplication
Principal Component Analysis (PCA) is a pivotal technique in the fields of machine learning and data analysis. It aims to reduce the dimensionality of a dataset while minimizing the loss of information. In recent years, there have been endeavors to utilize homomorphic encryption in privacy-preserving PCA algorithms. These approaches commonly employ a PCA routine known as PowerMethod, which takes the covariance matrix as input and generates an approximate eigenvector corresponding to the primary component of the dataset. However, their performance and accuracy are constrained by the incapability of homomorphic covariance matrix computation and the absence of a universal vector normalization strategy for the PowerMethod algorithm. In this study, we propose a novel approach to privacy-preserving PCA that addresses these limitations, resulting in superior efficiency, accuracy, and scalability compared to previous approaches. We attain such efficiency and precision through the following contributions: (i) We implement space optimization techniques for a homomorphic matrix multiplication method (Jiang et al., SIGSAC 2018), making it less prone to memory saturation in parallel computation scenarios. (ii) Leveraging the benefits of this optimized matrix multiplication, we devise an efficient homomorphic circuit for computing the covariance matrix homomorphically. (iii) Utilizing the covariance matrix, we develop a novel and efficient homomorphic circuit for the PowerMethod that incorporates a universal homomorphic vector normalization strategy to enhance both its accuracy and practicality.
♻ ☆ Blessings and Curses of Covariate Shifts: Adversarial Learning Dynamics, Directional Convergence, and Equilibria
Covariate distribution shifts and adversarial perturbations present robustness challenges to the conventional statistical learning framework: mild shifts in the test covariate distribution can significantly affect the performance of the statistical model learned based on the training distribution. The model performance typically deteriorates when extrapolation happens: namely, covariates shift to a region where the training distribution is scarce, and naturally, the learned model has little information. For robustness and regularization considerations, adversarial perturbation techniques are proposed as a remedy; however, careful study needs to be carried out about what extrapolation region adversarial covariate shift will focus on, given a learned model. This paper precisely characterizes the extrapolation region, examining both regression and classification in an infinite-dimensional setting. We study the implications of adversarial covariate shifts to subsequent learning of the equilibrium -- the Bayes optimal model -- in a sequential game framework. We exploit the dynamics of the adversarial learning game and reveal the curious effects of the covariate shift to equilibrium learning and experimental design. In particular, we establish two directional convergence results that exhibit distinctive phenomena: (1) a blessing in regression, the adversarial covariate shifts in an exponential rate to an optimal experimental design for rapid subsequent learning, (2) a curse in classification, the adversarial covariate shifts in a subquadratic rate fast to the hardest experimental design trapping subsequent learning.
comment: 22 pages, 2 figures
Multimedia 4
☆ An Overview of Challenges in Egocentric Text-Video Retrieval CVPR 2023
Text-video retrieval contains various challenges, including biases coming from diverse sources. We highlight some of them supported by illustrations to open a discussion. Besides, we address one of the biases, frame length bias, with a simple method which brings a very incremental but promising increase. We conclude with future directions.
comment: 4 pages, CVPR 2023 Joint Ego4D&EPIC Workshop, Extended Abstract
☆ Video Compression with Arbitrary Rescaling Network
Most video platforms provide video streaming services with different qualities, and the quality of the services is usually adjusted by the resolution of the videos. So high-resolution videos need to be downsampled for compression. In order to solve the problem of video coding at different resolutions, we propose a rate-guided arbitrary rescaling network (RARN) for video resizing before encoding. To help the RARN be compatible with standard codecs and generate compression-friendly results, an iteratively optimized transformer-based virtual codec (TVC) is introduced to simulate the key components of video encoding and perform bitrate estimation. By iteratively training the TVC and the RARN, we achieved 5%-29% BD-Rate reduction anchored by linear interpolation under different encoding configurations and resolutions, exceeding the previous methods on most test videos. Furthermore, the lightweight RARN structure can process FHD (1080p) content at real-time speed (91 FPS) and obtain a considerable rate reduction.
comment: Accepted as a one-page poster by 2023 Data Compression Conference (DCC). This is the full paper
♻ ☆ Digital Audio Forensics: Blind Human Voice Mimicry Detection
Audio is one of the most used ways of human communication, but at the same time it can be easily misused to trick people. With the revolution of AI, the related technologies are now accessible to almost everyone thus making it simple for the criminals to commit crimes and forgeries. In this work, we introduce a deep learning method to develop a classifier that will blindly classify an input audio as real or mimicked; the word 'blindly' refers to the ability to detect mimicked audio without references or real sources. The proposed model was trained on a set of important features extracted from a large dataset of audios to get a classifier that was tested on the same set of features from different audios. The data was extracted from two raw datasets, especially composed for this work; an all English dataset and a mixed dataset (Arabic plus English). These datasets have been made available, in raw form, through GitHub for the use of the research community at https://github.com/SaSs7/Dataset. For the purpose of comparison, the audios were also classified through human inspection with the subjects being the native speakers. The ensued results were interesting and exhibited formidable accuracy.
comment: 14 pages, 4 figures (6 if you count subfigures), 2 tables
♻ ☆ Occ-BEV: Multi-Camera Unified Pre-training via 3D Scene Reconstruction
Multi-camera 3D perception has emerged as a prominent research field in autonomous driving, offering a viable and cost-effective alternative to LiDAR-based solutions. However, existing multi-camera algorithms primarily rely on monocular image pre-training, which overlooks the spatial and temporal correlations among different camera views. To address this limitation, we propose the first multi-camera unified pre-training framework called Occ-BEV, which involves initially reconstructing the 3D scene as the foundational stage and subsequently fine-tuning the model on downstream tasks. Specifically, a 3D decoder is designed for leveraging Bird's Eye View (BEV) features from multi-view images to predict the 3D geometric occupancy to enable the model to capture a more comprehensive understanding of the 3D environment. A significant benefit of Occ-BEV is its capability of utilizing a considerable volume of unlabeled image-LiDAR pairs for pre-training purposes. The proposed multi-camera unified pre-training framework demonstrates promising results in key tasks such as multi-camera 3D object detection and surrounding semantic scene completion. When compared to monocular pre-training methods on the nuScenes dataset, Occ-BEV shows a significant improvement of about 2.0% in mAP and 2.0% in NDS for multi-camera 3D object detection, as well as a 3% increase in mIoU for surrounding semantic scene completion. Codes are publicly available at https://github.com/chaytonmin/Occ-BEV.
comment: 8 pages, 5 figures
Computation and Language 105
☆ CL-UZH at SemEval-2023 Task 10: Sexism Detection through Incremental Fine-Tuning and Multi-Task Learning with Label Descriptions ACL 2023
The widespread popularity of social media has led to an increase in hateful, abusive, and sexist language, motivating methods for the automatic detection of such phenomena. The goal of the SemEval shared task \textit{Towards Explainable Detection of Online Sexism} (EDOS 2023) is to detect sexism in English social media posts (subtask A), and to categorize such posts into four coarse-grained sexism categories (subtask B), and eleven fine-grained subcategories (subtask C). In this paper, we present our submitted systems for all three subtasks, based on a multi-task model that has been fine-tuned on a range of related tasks and datasets before being fine-tuned on the specific EDOS subtasks. We implement multi-task learning by formulating each task as binary pairwise text classification, where the dataset and label descriptions are given along with the input text. The results show clear improvements over a fine-tuned DeBERTa-V3 serving as a baseline leading to $F_1$-scores of 85.9\% in subtask A (rank 13/84), 64.8\% in subtask B (rank 19/69), and 44.9\% in subtask C (26/63).
comment: 11 pages, 4 figures, Accepted at The 17th International Workshop on Semantic Evaluation, ACL 2023
☆ Utterance Classification with Logical Neural Network: Explainable AI for Mental Disorder Diagnosis ACL 2023
In response to the global challenge of mental health problems, we proposes a Logical Neural Network (LNN) based Neuro-Symbolic AI method for the diagnosis of mental disorders. Due to the lack of effective therapy coverage for mental disorders, there is a need for an AI solution that can assist therapists with the diagnosis. However, current Neural Network models lack explainability and may not be trusted by therapists. The LNN is a Recurrent Neural Network architecture that combines the learning capabilities of neural networks with the reasoning capabilities of classical logic-based AI. The proposed system uses input predicates from clinical interviews to output a mental disorder class, and different predicate pruning techniques are used to achieve scalability and higher scores. In addition, we provide an insight extraction method to aid therapists with their diagnosis. The proposed system addresses the lack of explainability of current Neural Network models and provides a more trustworthy solution for mental disorder diagnosis.
comment: ACL 2023
☆ ChatDB: Augmenting LLMs with Databases as Their Symbolic Memory
Large language models (LLMs) with memory are computationally universal. However, mainstream LLMs are not taking full advantage of memory, and the designs are heavily influenced by biological brains. Due to their approximate nature and proneness to the accumulation of errors, conventional neural memory mechanisms cannot support LLMs to simulate complex reasoning. In this paper, we seek inspiration from modern computer architectures to augment LLMs with symbolic memory for complex multi-hop reasoning. Such a symbolic memory framework is instantiated as an LLM and a set of SQL databases, where the LLM generates SQL instructions to manipulate the SQL databases. We validate the effectiveness of the proposed memory framework on a synthetic dataset requiring complex reasoning. The project website is available at https://chatdatabase.github.io/ .
☆ Causal interventions expose implicit situation models for commonsense language understanding ACL
Accounts of human language processing have long appealed to implicit ``situation models'' that enrich comprehension with relevant but unstated world knowledge. Here, we apply causal intervention techniques to recent transformer models to analyze performance on the Winograd Schema Challenge (WSC), where a single context cue shifts interpretation of an ambiguous pronoun. We identify a relatively small circuit of attention heads that are responsible for propagating information from the context word that guides which of the candidate noun phrases the pronoun ultimately attends to. We then compare how this circuit behaves in a closely matched ``syntactic'' control where the situation model is not strictly necessary. These analyses suggest distinct pathways through which implicit situation models are constructed to guide pronoun resolution.
comment: Findings of ACL
☆ Deductive Verification of Chain-of-Thought Reasoning
Large Language Models (LLMs) significantly benefit from Chain-of-Thought (CoT) prompting in performing various reasoning tasks. While CoT allows models to produce more comprehensive reasoning processes, its emphasis on intermediate reasoning steps can inadvertently introduce hallucinations and accumulated errors, thereby limiting models' ability to solve complex reasoning tasks. Inspired by how humans engage in careful and meticulous deductive logical reasoning processes to solve tasks, we seek to enable language models to perform explicit and rigorous deductive reasoning, and also ensure the trustworthiness of their reasoning process through self-verification. However, directly verifying the validity of an entire deductive reasoning process is challenging, even with advanced models like ChatGPT. In light of this, we propose to decompose a reasoning verification process into a series of step-by-step subprocesses, each only receiving their necessary context and premises. To facilitate this procedure, we propose Natural Program, a natural language-based deductive reasoning format. Our approach enables models to generate precise reasoning steps where subsequent steps are more rigorously grounded on prior steps. It also empowers language models to carry out reasoning self-verification in a step-by-step manner. By integrating this verification process into each deductive reasoning stage, we significantly enhance the rigor and trustfulness of generated reasoning steps. Along this process, we also improve the answer correctness on complex reasoning tasks. Code will be released at https://github.com/lz1oceani/verify_cot.
☆ Correction of Errors in Preference Ratings from Automated Metrics for Text Generation
A major challenge in the field of Text Generation is evaluation: Human evaluations are cost-intensive, and automated metrics often display considerable disagreement with human judgments. In this paper, we propose a statistical model of Text Generation evaluation that accounts for the error-proneness of automated metrics when used to generate preference rankings between system outputs. We show that existing automated metrics are generally over-confident in assigning significant differences between systems in this setting. However, our model enables an efficient combination of human and automated ratings to remedy the error-proneness of the automated metrics. We show that using this combination, we only require about 50% of the human annotations typically used in evaluations to arrive at robust and statistically significant results while yielding the same evaluation outcome as the pure human evaluation in 95% of cases. We showcase the benefits of approach for three text generation tasks: dialogue systems, machine translation, and text summarization.
☆ Iterative Translation Refinement with Large Language Models
Large language models have shown surprising performances in understanding instructions and performing natural language tasks. In this paper, we propose iterative translation refinement to leverage the power of large language models for more natural translation and post-editing. We show that by simply involving a large language model in an iterative process, the output quality improves beyond mere translation. Extensive test scenarios with GPT-3.5 reveal that although iterations reduce string-based metric scores, neural metrics indicate comparable if not improved translation quality. Further, human evaluations demonstrate that our method effectively reduces translationese compared to initial GPT translations and even human references, especially for into-English directions. Ablation studies underscore the importance of anchoring the refinement process to the source input and a reasonable initial translation.
☆ From Key Points to Key Point Hierarchy: Structured and Expressive Opinion Summarization ACL 2023
Key Point Analysis (KPA) has been recently proposed for deriving fine-grained insights from collections of textual comments. KPA extracts the main points in the data as a list of concise sentences or phrases, termed key points, and quantifies their prevalence. While key points are more expressive than word clouds and key phrases, making sense of a long, flat list of key points, which often express related ideas in varying levels of granularity, may still be challenging. To address this limitation of KPA, we introduce the task of organizing a given set of key points into a hierarchy, according to their specificity. Such hierarchies may be viewed as a novel type of Textual Entailment Graph. We develop ThinkP, a high quality benchmark dataset of key point hierarchies for business and product reviews, obtained by consolidating multiple annotations. We compare different methods for predicting pairwise relations between key points, and for inferring a hierarchy from these pairwise predictions. In particular, for the task of computing pairwise key point relations, we achieve significant gains over existing strong baselines by applying directional distributional similarity methods to a novel distributional representation of key points, and further boost performance via weak supervision.
comment: ACL 2023
☆ LEACE: Perfect linear concept erasure in closed form
Concept erasure aims to remove specified features from a representation. It can be used to improve fairness (e.g. preventing a classifier from using gender or race) and interpretability (e.g. removing a concept to observe changes in model behavior). In this paper, we introduce LEAst-squares Concept Erasure (LEACE), a closed-form method which provably prevents all linear classifiers from detecting a concept while inflicting the least possible damage to the representation. We apply LEACE to large language models with a novel procedure called "concept scrubbing," which erases target concept information from every layer in the network. We demonstrate the usefulness of our method on two tasks: measuring the reliance of language models on part-of-speech information, and reducing gender bias in BERT embeddings. Code is available at https://github.com/EleutherAI/concept-erasure.
Prompt Space Optimizing Few-shot Reasoning Success with Large Language Models
Prompt engineering is an essential technique for enhancing the abilities of large language models (LLMs) by providing explicit and specific instructions. It enables LLMs to excel in various tasks, such as arithmetic reasoning, question answering, summarization, relation extraction, machine translation, and sentiment analysis. Researchers have been actively exploring different prompt engineering strategies, such as Chain of Thought (CoT), Zero-CoT, and In-context learning. However, an unresolved problem arises from the fact that current approaches lack a solid theoretical foundation for determining optimal prompts. To address this issue in prompt engineering, we propose a new and effective approach called Prompt Space. Our methodology utilizes text embeddings to obtain basis vectors by matrix decomposition, and then constructs a space for representing all prompts. Prompt Space significantly outperforms state-of-the-art prompt paradigms on ten public reasoning benchmarks. Notably, without the help of the CoT method and the prompt "Let's think step by step", Prompt Space shows superior performance over the few-shot method. Overall, our approach provides a robust and fundamental theoretical framework for selecting simple and effective prompts. This advancement marks a significant step towards improving prompt engineering for a wide variety of applications in LLMs.
comment: Natural language processing (NLP)
☆ Exploring Linguistic Features for Turkish Text Readability
This paper presents the first comprehensive study on automatic readability assessment of Turkish texts. We combine state-of-the-art neural network models with linguistic features at lexical, morphosyntactic, syntactic and discourse levels to develop an advanced readability tool. We evaluate the effectiveness of traditional readability formulas compared to modern automated methods and identify key linguistic features that determine the readability of Turkish texts.
☆ FinRED: A Dataset for Relation Extraction in Financial Domain WWW'22
Relation extraction models trained on a source domain cannot be applied on a different target domain due to the mismatch between relation sets. In the current literature, there is no extensive open-source relation extraction dataset specific to the finance domain. In this paper, we release FinRED, a relation extraction dataset curated from financial news and earning call transcripts containing relations from the finance domain. FinRED has been created by mapping Wikidata triplets using distance supervision method. We manually annotate the test data to ensure proper evaluation. We also experiment with various state-of-the-art relation extraction models on this dataset to create the benchmark. We see a significant drop in their performance on FinRED compared to the general relation extraction datasets which tells that we need better models for financial relation extraction.
comment: Accepted at FinWeb at WWW'22
☆ A Cross-Linguistic Pressure for Uniform Information Density in Word Order
While natural languages differ widely in both canonical word order and word order flexibility, their word orders still follow shared cross-linguistic statistical patterns, often attributed to functional pressures. In the effort to identify these pressures, prior work has compared real and counterfactual word orders. Yet one functional pressure has been overlooked in such investigations: the uniform information density (UID) hypothesis, which holds that information should be spread evenly throughout an utterance. Here, we ask whether a pressure for UID may have influenced word order patterns cross-linguistically. To this end, we use computational models to test whether real orders lead to greater information uniformity than counterfactual orders. In our empirical study of 10 typologically diverse languages, we find that: (i) among SVO languages, real word orders consistently have greater uniformity than reverse word orders, and (ii) only linguistically implausible counterfactual orders consistently exceed the uniformity of real orders. These findings are compatible with a pressure for information uniformity in the development and usage of natural languages.
☆ A Novel Approach To User Agent String Parsing For Vulnerability Analysis Using Mutli-Headed Attention ICML
The increasing reliance on the internet has led to the proliferation of a diverse set of web-browsers and operating systems (OSs) capable of browsing the web. User agent strings (UASs) are a component of web browsing that are transmitted with every Hypertext Transfer Protocol (HTTP) request. They contain information about the client device and software, which is used by web servers for various purposes such as content negotiation and security. However, due to the proliferation of various browsers and devices, parsing UASs is a non-trivial task due to a lack of standardization of UAS formats. Current rules-based approaches are often brittle and can fail when encountering such non-standard formats. In this work, a novel methodology for parsing UASs using Multi-Headed Attention Based transformers is proposed. The proposed methodology exhibits strong performance in parsing a variety of UASs with differing formats. Furthermore, a framework to utilize parsed UASs to estimate the vulnerability scores for large sections of publicly visible IT networks or regions is also discussed. The methodology present here can also be easily extended or deployed for real-time parsing of logs in enterprise settings.
comment: Accepted to the International Conference on Machine Learning and Cybernetics (ICMLC) 2023
☆ Financial Numeric Extreme Labelling: A Dataset and Benchmarking for XBRL Tagging ACL'23
The U.S. Securities and Exchange Commission (SEC) mandates all public companies to file periodic financial statements that should contain numerals annotated with a particular label from a taxonomy. In this paper, we formulate the task of automating the assignment of a label to a particular numeral span in a sentence from an extremely large label set. Towards this task, we release a dataset, Financial Numeric Extreme Labelling (FNXL), annotated with 2,794 labels. We benchmark the performance of the FNXL dataset by formulating the task as (a) a sequence labelling problem and (b) a pipeline with span extraction followed by Extreme Classification. Although the two approaches perform comparably, the pipeline solution provides a slight edge for the least frequent labels.
comment: Accepted to ACL'23 Findings Paper
☆ Evaluating the Effectiveness of Natural Language Inference for Hate Speech Detection in Languages with Limited Labeled Data WOAH
Most research on hate speech detection has focused on English where a sizeable amount of labeled training data is available. However, to expand hate speech detection into more languages, approaches that require minimal training data are needed. In this paper, we test whether natural language inference (NLI) models which perform well in zero- and few-shot settings can benefit hate speech detection performance in scenarios where only a limited amount of labeled data is available in the target language. Our evaluation on five languages demonstrates large performance improvements of NLI fine-tuning over direct fine-tuning in the target language. However, the effectiveness of previous work that proposed intermediate fine-tuning on English data is hard to match. Only in settings where the English training data does not match the test domain, can our customised NLI-formulation outperform intermediate fine-tuning on English. Based on our extensive experiments, we propose a set of recommendations for hate speech detection in languages where minimal labeled training data is available.
comment: 15 pages, 7 figures, Accepted at the 7th Workshop on Online Abuse and Harms (WOAH), ACL 2023
☆ On the Difference of BERT-style and CLIP-style Text Encoders ACL-2023
Masked language modeling (MLM) has been one of the most popular pretraining recipes in natural language processing, e.g., BERT, one of the representative models. Recently, contrastive language-image pretraining (CLIP) has also attracted attention, especially its vision models that achieve excellent performance on a broad range of vision tasks. However, few studies are dedicated to studying the text encoders learned by CLIP. In this paper, we analyze the difference between BERT-style and CLIP-style text encoders from three experiments: (i) general text understanding, (ii) vision-centric text understanding, and (iii) text-to-image generation. Experimental analyses show that although CLIP-style text encoders underperform BERT-style ones for general text understanding tasks, they are equipped with a unique ability, i.e., synesthesia, for the cross-modal association, which is more similar to the senses of humans.
comment: Natural Language Processing. 10 pages, 1 figure. Findings of ACL-2023
☆ Injecting knowledge into language generation: a case study in auto-charting after-visit care instructions from medical dialogue ACL 2023
Factual correctness is often the limiting factor in practical applications of natural language generation in high-stakes domains such as healthcare. An essential requirement for maintaining factuality is the ability to deal with rare tokens. This paper focuses on rare tokens that appear in both the source and the reference sequences, and which, when missed during generation, decrease the factual correctness of the output text. For high-stake domains that are also knowledge-rich, we show how to use knowledge to (a) identify which rare tokens that appear in both source and reference are important and (b) uplift their conditional probability. We introduce the ``utilization rate'' that encodes knowledge and serves as a regularizer by maximizing the marginal probability of selected tokens. We present a study in a knowledge-rich domain of healthcare, where we tackle the problem of generating after-visit care instructions based on patient-doctor dialogues. We verify that, in our dataset, specific medical concepts with high utilization rates are underestimated by conventionally trained sequence-to-sequence models. We observe that correcting this with our approach to knowledge injection reduces the uncertainty of the model as well as improves factuality and coherence without negatively impacting fluency.
comment: ACL 2023 (main conference)
☆ A Quantum Probability Driven Framework for Joint Multi-Modal Sarcasm, Sentiment and Emotion Analysis
Sarcasm, sentiment, and emotion are three typical kinds of spontaneous affective responses of humans to external events and they are tightly intertwined with each other. Such events may be expressed in multiple modalities (e.g., linguistic, visual and acoustic), e.g., multi-modal conversations. Joint analysis of humans' multi-modal sarcasm, sentiment, and emotion is an important yet challenging topic, as it is a complex cognitive process involving both cross-modality interaction and cross-affection correlation. From the probability theory perspective, cross-affection correlation also means that the judgments on sarcasm, sentiment, and emotion are incompatible. However, this exposed phenomenon cannot be sufficiently modelled by classical probability theory due to its assumption of compatibility. Neither do the existing approaches take it into consideration. In view of the recent success of quantum probability (QP) in modeling human cognition, particularly contextual incompatible decision making, we take the first step towards introducing QP into joint multi-modal sarcasm, sentiment, and emotion analysis. Specifically, we propose a QUantum probabIlity driven multi-modal sarcasm, sEntiment and emoTion analysis framework, termed QUIET. Extensive experiments on two datasets and the results show that the effectiveness and advantages of QUIET in comparison with a wide range of the state-of-the-art baselines. We also show the great potential of QP in multi-affect analysis.
☆ Convergence and Diversity in the Control Hierarchy ACL 2023
Weir has defined a hierarchy of language classes whose second member ($\mathcal{L}_2$) is generated by tree-adjoining grammars (TAG), linear indexed grammars (LIG), combinatory categorial grammars, and head grammars. The hierarchy is obtained using the mechanism of control, and $\mathcal{L}_2$ is obtained using a context-free grammar (CFG) whose derivations are controlled by another CFG. We adapt Weir's definition of a controllable CFG to give a definition of controllable pushdown automata (PDAs). This yields three new characterizations of $\mathcal{L}_2$ as the class of languages generated by PDAs controlling PDAs, PDAs controlling CFGs, and CFGs controlling PDAs. We show that these four formalisms are not only weakly equivalent but equivalent in a stricter sense that we call d-weak equivalence. Furthermore, using an even stricter notion of equivalence called d-strong equivalence, we make precise the intuition that a CFG controlling a CFG is a TAG, a PDA controlling a PDA is an embedded PDA, and a PDA controlling a CFG is a LIG. The fourth member of this family, a CFG controlling a PDA, does not correspond to any formalism we know of, so we invent one and call it a Pushdown Adjoining Automaton.
comment: 18 pages, 7 figures. Accepted at ACL 2023
☆ A Survey of Quantum-Cognitively Inspired Sentiment Analysis Models
Quantum theory, originally proposed as a physical theory to describe the motions of microscopic particles, has been applied to various non-physics domains involving human cognition and decision-making that are inherently uncertain and exhibit certain non-classical, quantum-like characteristics. Sentiment analysis is a typical example of such domains. In the last few years, by leveraging the modeling power of quantum probability (a non-classical probability stemming from quantum mechanics methodology) and deep neural networks, a range of novel quantum-cognitively inspired models for sentiment analysis have emerged and performed well. This survey presents a timely overview of the latest developments in this fascinating cross-disciplinary area. We first provide a background of quantum probability and quantum cognition at a theoretical level, analyzing their advantages over classical theories in modeling the cognitive aspects of sentiment analysis. Then, recent quantum-cognitively inspired models are introduced and discussed in detail, focusing on how they approach the key challenges of the sentiment analysis task. Finally, we discuss the limitations of the current research and highlight future research directions.
☆ CUE: An Uncertainty Interpretation Framework for Text Classifiers Built on Pre-Trained Language Models UAI 2023
Text classifiers built on Pre-trained Language Models (PLMs) have achieved remarkable progress in various tasks including sentiment analysis, natural language inference, and question-answering. However, the occurrence of uncertain predictions by these classifiers poses a challenge to their reliability when deployed in practical applications. Much effort has been devoted to designing various probes in order to understand what PLMs capture. But few studies have delved into factors influencing PLM-based classifiers' predictive uncertainty. In this paper, we propose a novel framework, called CUE, which aims to interpret uncertainties inherent in the predictions of PLM-based models. In particular, we first map PLM-encoded representations to a latent space via a variational auto-encoder. We then generate text representations by perturbing the latent space which causes fluctuation in predictive uncertainty. By comparing the difference in predictive uncertainty between the perturbed and the original text representations, we are able to identify the latent dimensions responsible for uncertainty and subsequently trace back to the input features that contribute to such uncertainty. Our extensive experiments on four benchmark datasets encompassing linguistic acceptability classification, emotion classification, and natural language inference show the feasibility of our proposed framework. Our source code is available at: https://github.com/lijiazheng99/CUE.
comment: Accepted to UAI 2023
☆ Language acquisition: do children and language models follow similar learning stages? ACL 2023
During language acquisition, children follow a typical sequence of learning stages, whereby they first learn to categorize phonemes before they develop their lexicon and eventually master increasingly complex syntactic structures. However, the computational principles that lead to this learning trajectory remain largely unknown. To investigate this, we here compare the learning trajectories of deep language models to those of children. Specifically, we test whether, during its training, GPT-2 exhibits stages of language acquisition comparable to those observed in children aged between 18 months and 6 years. For this, we train 48 GPT-2 models from scratch and evaluate their syntactic and semantic abilities at each training step, using 96 probes curated from the BLiMP, Zorro and BIG-Bench benchmarks. We then compare these evaluations with the behavior of 54 children during language production. Our analyses reveal three main findings. First, similarly to children, the language models tend to learn linguistic skills in a systematic order. Second, this learning scheme is parallel: the language tasks that are learned last improve from the very first training steps. Third, some - but not all - learning stages are shared between children and these language models. Overall, these results shed new light on the principles of language acquisition, and highlight important divergences in how humans and modern algorithms learn to process natural language.
comment: Accepted to ACL 2023. *Equal Contribution
☆ Take the Hint: Improving Arabic Diacritization with Partially-Diacritized Text
Automatic Arabic diacritization is useful in many applications, ranging from reading support for language learners to accurate pronunciation predictor for downstream tasks like speech synthesis. While most of the previous works focused on models that operate on raw non-diacritized text, production systems can gain accuracy by first letting humans partly annotate ambiguous words. In this paper, we propose 2SDiac, a multi-source model that can effectively support optional diacritics in input to inform all predictions. We also introduce Guided Learning, a training scheme to leverage given diacritics in input with different levels of random masking. We show that the provided hints during test affect more output positions than those annotated. Moreover, experiments on two common benchmarks show that our approach i) greatly outperforms the baseline also when evaluated on non-diacritized text; and ii) achieves state-of-the-art results while reducing the parameter count by over 60%.
comment: Arabic text diacritization, partially-diacritized text, Arabic natural language processing
☆ SciLit: A Platform for Joint Scientific Literature Discovery, Summarization and Citation Generation ACL 2023
Scientific writing involves retrieving, summarizing, and citing relevant papers, which can be time-consuming processes in large and rapidly evolving fields. By making these processes inter-operable, natural language processing (NLP) provides opportunities for creating end-to-end assistive writing tools. We propose SciLit, a pipeline that automatically recommends relevant papers, extracts highlights, and suggests a reference sentence as a citation of a paper, taking into consideration the user-provided context and keywords. SciLit efficiently recommends papers from large databases of hundreds of millions of papers using a two-stage pre-fetching and re-ranking literature search system that flexibly deals with addition and removal of a paper database. We provide a convenient user interface that displays the recommended papers as extractive summaries and that offers abstractively-generated citing sentences which are aligned with the provided context and which mention the chosen keyword(s). Our assistive tool for literature discovery and scientific writing is available at https://scilit.vercel.app
comment: Accepted at ACL 2023 System Demonstration
☆ "A Little is Enough": Few-Shot Quality Estimation based Corpus Filtering improves Machine Translation
Quality Estimation (QE) is the task of evaluating the quality of a translation when reference translation is not available. The goal of QE aligns with the task of corpus filtering, where we assign the quality score to the sentence pairs present in the pseudo-parallel corpus. We propose a Quality Estimation based Filtering approach to extract high-quality parallel data from the pseudo-parallel corpus. To the best of our knowledge, this is a novel adaptation of the QE framework to extract quality parallel corpus from the pseudo-parallel corpus. By training with this filtered corpus, we observe an improvement in the Machine Translation (MT) system's performance by up to 1.8 BLEU points, for English-Marathi, Chinese-English, and Hindi-Bengali language pairs, over the baseline model. The baseline model is the one that is trained on the whole pseudo-parallel corpus. Our Few-shot QE model transfer learned from the English-Marathi QE model and fine-tuned on only 500 Hindi-Bengali training instances, shows an improvement of up to 0.6 BLEU points for Hindi-Bengali language pair, compared to the baseline model. This demonstrates the promise of transfer learning in the setting under discussion. QE systems typically require in the order of (7K-25K) of training data. Our Hindi-Bengali QE is trained on only 500 instances of training that is 1/40th of the normal requirement and achieves comparable performance. All the scripts and datasets utilized in this study will be publicly available.
☆ Applying Standards to Advance Upstream & Downstream Ethics in Large Language Models
This paper explores how AI-owners can develop safeguards for AI-generated content by drawing from established codes of conduct and ethical standards in other content-creation industries. It delves into the current state of ethical awareness on Large Language Models (LLMs). By dissecting the mechanism of content generation by LLMs, four key areas (upstream/downstream and at user prompt/answer), where safeguards could be effectively applied, are identified. A comparative analysis of these four areas follows and includes an evaluation of the existing ethical safeguards in terms of cost, effectiveness, and alignment with established industry practices. The paper's key argument is that existing IT-related ethical codes, while adequate for traditional IT engineering, are inadequate for the challenges posed by LLM-based content generation. Drawing from established practices within journalism, we propose potential standards for businesses involved in distributing and selling LLM-generated content. Finally, potential conflicts of interest between dataset curation at upstream and ethical benchmarking downstream are highlighted to underscore the need for a broader evaluation beyond mere output. This study prompts a nuanced conversation around ethical implications in this rapidly evolving field of content generation.
comment: 8 pages, 4 tables, 2 figures
☆ Towards Adaptable and Interactive Image Captioning with Data Augmentation and Episodic Memory
Interactive machine learning (IML) is a beneficial learning paradigm in cases of limited data availability, as human feedback is incrementally integrated into the training process. In this paper, we present an IML pipeline for image captioning which allows us to incrementally adapt a pre-trained image captioning model to a new data distribution based on user input. In order to incorporate user input into the model, we explore the use of a combination of simple data augmentation methods to obtain larger data batches for each newly annotated data instance and implement continual learning methods to prevent catastrophic forgetting from repeated updates. For our experiments, we split a domain-specific image captioning dataset, namely VizWiz, into non-overlapping parts to simulate an incremental input flow for continually adapting the model to new data. We find that, while data augmentation worsens results, even when relatively small amounts of data are available, episodic memory is an effective strategy to retain knowledge from previously seen clusters.
☆ SciCap+: A Knowledge Augmented Dataset to Study the Challenges of Scientific Figure Captioning AAAI23
In scholarly documents, figures provide a straightforward way of communicating scientific findings to readers. Automating figure caption generation helps move model understandings of scientific documents beyond text and will help authors write informative captions that facilitate communicating scientific findings. Unlike previous studies, we reframe scientific figure captioning as a knowledge-augmented image captioning task that models need to utilize knowledge embedded across modalities for caption generation. To this end, we extended the large-scale SciCap dataset~\cite{hsu-etal-2021-scicap-generating} to SciCap+ which includes mention-paragraphs (paragraphs mentioning figures) and OCR tokens. Then, we conduct experiments with the M4C-Captioner (a multimodal transformer-based model with a pointer network) as a baseline for our study. Our results indicate that mention-paragraphs serves as additional context knowledge, which significantly boosts the automatic standard image caption evaluation scores compared to the figure-only baselines. Human evaluations further reveal the challenges of generating figure captions that are informative to readers. The code and SciCap+ dataset will be publicly available at https://github.com/ZhishenYang/scientific_figure_captioning_dataset
comment: Published in SDU workshop at AAAI23
☆ Putting Humans in the Image Captioning Loop
Image Captioning (IC) models can highly benefit from human feedback in the training process, especially in cases where data is limited. We present work-in-progress on adapting an IC system to integrate human feedback, with the goal to make it easily adaptable to user-specific data. Our approach builds on a base IC model pre-trained on the MS COCO dataset, which generates captions for unseen images. The user will then be able to offer feedback on the image and the generated/predicted caption, which will be augmented to create additional training instances for the adaptation of the model. The additional instances are integrated into the model using step-wise updates, and a sparse memory replay component is used to avoid catastrophic forgetting. We hope that this approach, while leading to improved results, will also result in customizable IC models.
☆ Joint Event Extraction via Structural Semantic Matching
Event Extraction (EE) is one of the essential tasks in information extraction, which aims to detect event mentions from text and find the corresponding argument roles. The EE task can be abstracted as a process of matching the semantic definitions and argument structures of event types with the target text. This paper encodes the semantic features of event types and makes structural matching with target text. Specifically, Semantic Type Embedding (STE) and Dynamic Structure Encoder (DSE) modules are proposed. Also, the Joint Structural Semantic Matching (JSSM) model is built to jointly perform event detection and argument extraction tasks through a bidirectional attention layer. The experimental results on the ACE2005 dataset indicate that our model achieves a significant performance improvement
☆ Natural Language Commanding via Program Synthesis
We present Semantic Interpreter, a natural language-friendly AI system for productivity software such as Microsoft Office that leverages large language models (LLMs) to execute user intent across application features. While LLMs are excellent at understanding user intent expressed as natural language, they are not sufficient for fulfilling application-specific user intent that requires more than text-to-text transformations. We therefore introduce the Office Domain Specific Language (ODSL), a concise, high-level language specialized for performing actions in and interacting with entities in Office applications. Semantic Interpreter leverages an Analysis-Retrieval prompt construction method with LLMs for program synthesis, translating natural language user utterances to ODSL programs that can be transpiled to application APIs and then executed. We focus our discussion primarily on a research exploration for Microsoft PowerPoint.
☆ Phonetically-Grounded Language Generation: The Case of Tongue Twisters
Previous work in phonetically-grounded language generation has mainly focused on domains such as lyrics and poetry. In this paper, we present work on the generation of tongue twisters - a form of language that is required to be phonetically conditioned to maximise sound overlap, whilst maintaining semantic consistency with an input topic, and still being grammatically correct. We present \textbf{TwistList}, a large annotated dataset of tongue twisters, consisting of 2.1K+ human-authored examples. We additionally present several benchmark systems (referred to as TwisterMisters) for the proposed task of tongue twister generation, including models that both do and do not require training on in-domain data. We present the results of automatic and human evaluation to demonstrate the performance of existing mainstream pre-trained models in this task with limited (or no) task specific training and data, and no explicit phonetic knowledge. We find that the task of tongue twister generation is challenging for models under these conditions, yet some models are still capable of generating acceptable examples of this language type.
☆ Automatic Assessment of Oral Reading Accuracy for Reading Diagnostics
Automatic assessment of reading fluency using automatic speech recognition (ASR) holds great potential for early detection of reading difficulties and subsequent timely intervention. Precise assessment tools are required, especially for languages other than English. In this study, we evaluate six state-of-the-art ASR-based systems for automatically assessing Dutch oral reading accuracy using Kaldi and Whisper. Results show our most successful system reached substantial agreement with human evaluations (MCC = .63). The same system reached the highest correlation between forced decoding confidence scores and word correctness (r = .45). This system's language model (LM) consisted of manual orthographic transcriptions and reading prompts of the test data, which shows that including reading errors in the LM improves assessment performance. We discuss the implications for developing automatic assessment systems and identify possible avenues of future research.
☆ Alzheimer Disease Classification through ASR-based Transcriptions: Exploring the Impact of Punctuation and Pauses
Alzheimer's Disease (AD) is the world's leading neurodegenerative disease, which often results in communication difficulties. Analysing speech can serve as a diagnostic tool for identifying the condition. The recent ADReSS challenge provided a dataset for AD classification and highlighted the utility of manual transcriptions. In this study, we used the new state-of-the-art Automatic Speech Recognition (ASR) model Whisper to obtain the transcriptions, which also include automatic punctuation. The classification models achieved test accuracy scores of 0.854 and 0.833 combining the pretrained FastText word embeddings and recurrent neural networks on manual and ASR transcripts respectively. Additionally, we explored the influence of including pause information and punctuation in the transcriptions. We found that punctuation only yielded minor improvements in some cases, whereas pause encoding aided AD classification for both manual and ASR transcriptions across all approaches investigated.
☆ Large Language Models of Code Fail at Completing Code with Potential Bugs
Large language models of code (Code-LLMs) have recently brought tremendous advances to code completion, a fundamental feature of programming assistance and code intelligence. However, most existing works ignore the possible presence of bugs in the code context for generation, which are inevitable in software development. Therefore, we introduce and study the buggy-code completion problem, inspired by the realistic scenario of real-time code suggestion where the code context contains potential bugs -- anti-patterns that can become bugs in the completed program. To systematically study the task, we introduce two datasets: one with synthetic bugs derived from semantics-altering operator changes (buggy-HumanEval) and one with realistic bugs derived from user submissions to coding problems (buggy-FixEval). We find that the presence of potential bugs significantly degrades the generation performance of the high-performing Code-LLMs. For instance, the passing rates of CodeGen-2B-mono on test cases of buggy-HumanEval drop more than 50% given a single potential bug in the context. Finally, we investigate several post-hoc methods for mitigating the adverse effect of potential bugs and find that there remains a large gap in post-mitigation performance.
comment: 25 pages
☆ On the Role of Attention in Prompt-tuning ICML 2023
Prompt-tuning is an emerging strategy to adapt large language models (LLM) to downstream tasks by learning a (soft-)prompt parameter from data. Despite its success in LLMs, there is limited theoretical understanding of the power of prompt-tuning and the role of the attention mechanism in prompting. In this work, we explore prompt-tuning for one-layer attention architectures and study contextual mixture-models where each input token belongs to a context-relevant or -irrelevant set. We isolate the role of prompt-tuning through a self-contained prompt-attention model. Our contributions are as follows: (1) We show that softmax-prompt-attention is provably more expressive than softmax-self-attention and linear-prompt-attention under our contextual data model. (2) We analyze the initial trajectory of gradient descent and show that it learns the prompt and prediction head with near-optimal sample complexity and demonstrate how prompt can provably attend to sparse context-relevant tokens. (3) Assuming a known prompt but an unknown prediction head, we characterize the exact finite sample performance of prompt-attention which reveals the fundamental performance limits and the precise benefit of the context information. We also provide experiments that verify our theoretical insights on real datasets and demonstrate how prompt-tuning enables the model to attend to context-relevant information.
comment: Published at ICML 2023
☆ Efficient and Interpretable Compressive Text Summarisation with Unsupervised Dual-Agent Reinforcement Learning ACL 2023
Recently, compressive text summarisation offers a balance between the conciseness issue of extractive summarisation and the factual hallucination issue of abstractive summarisation. However, most existing compressive summarisation methods are supervised, relying on the expensive effort of creating a new training dataset with corresponding compressive summaries. In this paper, we propose an efficient and interpretable compressive summarisation method that utilises unsupervised dual-agent reinforcement learning to optimise a summary's semantic coverage and fluency by simulating human judgment on summarisation quality. Our model consists of an extractor agent and a compressor agent, and both agents have a multi-head attentional pointer-based structure. The extractor agent first chooses salient sentences from a document, and then the compressor agent compresses these extracted sentences by selecting salient words to form a summary without using reference summaries to compute the summary reward. To our best knowledge, this is the first work on unsupervised compressive summarisation. Experimental results on three widely used datasets (e.g., Newsroom, CNN/DM, and XSum) show that our model achieves promising performance and a significant improvement on Newsroom in terms of the ROUGE metric, as well as interpretability of semantic coverage of summarisation results.
comment: The 4th Workshop on Simple and Efficient Natural Language Processing (SustaiNLP 2023), co-located with ACL 2023
☆ Generate-then-Retrieve: Intent-Aware FAQ Retrieval in Product Search ACL 2023
Customers interacting with product search engines are increasingly formulating information-seeking queries. Frequently Asked Question (FAQ) retrieval aims to retrieve common question-answer pairs for a user query with question intent. Integrating FAQ retrieval in product search can not only empower users to make more informed purchase decisions, but also enhance user retention through efficient post-purchase support. Determining when an FAQ entry can satisfy a user's information need within product search, without disrupting their shopping experience, represents an important challenge. We propose an intent-aware FAQ retrieval system consisting of (1) an intent classifier that predicts when a user's information need can be answered by an FAQ; (2) a reformulation model that rewrites a query into a natural question. Offline evaluation demonstrates that our approach improves Hit@1 by 13% on retrieving ground-truth FAQs, while reducing latency by 95% compared to baseline systems. These improvements are further validated by real user feedback, where 71% of displayed FAQs on top of product search results received explicit positive user feedback. Overall, our findings show promising directions for integrating FAQ retrieval into product search at scale.
comment: ACL 2023 Industry Track
☆ TextFormer: A Query-based End-to-End Text Spotter with Mixed Supervision
End-to-end text spotting is a vital computer vision task that aims to integrate scene text detection and recognition into a unified framework. Typical methods heavily rely on Region-of-Interest (RoI) operations to extract local features and complex post-processing steps to produce final predictions. To address these limitations, we propose TextFormer, a query-based end-to-end text spotter with Transformer architecture. Specifically, using query embedding per text instance, TextFormer builds upon an image encoder and a text decoder to learn a joint semantic understanding for multi-task modeling. It allows for mutual training and optimization of classification, segmentation, and recognition branches, resulting in deeper feature sharing without sacrificing flexibility or simplicity. Additionally, we design an Adaptive Global aGgregation (AGG) module to transfer global features into sequential features for reading arbitrarily-shaped texts, which overcomes the sub-optimization problem of RoI operations. Furthermore, potential corpus information is utilized from weak annotations to full labels through mixed supervision, further improving text detection and end-to-end text spotting results. Extensive experiments on various bilingual (i.e., English and Chinese) benchmarks demonstrate the superiority of our method. Especially on TDA-ReCTS dataset, TextFormer surpasses the state-of-the-art method in terms of 1-NED by 13.2%.
comment: MIR 2023, 15 pages
☆ $\textit{WHAT}$, $\textit{WHEN}$, and $\textit{HOW}$ to Ground: Designing User Persona-Aware Conversational Agents for Engaging Dialogue ACL 2023
This paper presents a method for building a personalized open-domain dialogue system to address the $\textit{WWH}$ ($\textit{WHAT}$, $\textit{WHEN}$, and $\textit{HOW}$) problem for natural response generation in a commercial setting, where personalized dialogue responses are heavily interleaved with casual response turns. The proposed approach involves weighted dataset blending, negative persona information augmentation methods, and the design of personalized conversation datasets to address the challenges of $\textit{WWH}$ in personalized, open-domain dialogue systems. Our work effectively balances dialogue fluency and tendency to ground, while also introducing a response-type label to improve the controllability and explainability of the grounded responses. The combination of these methods leads to more fluent conversations, as evidenced by subjective human evaluations as well as objective evaluations.
comment: Accepted in ACL 2023 Industry Track
☆ BatchSampler: Sampling Mini-Batches for Contrastive Learning in Vision, Language, and Graphs
In-Batch contrastive learning is a state-of-the-art self-supervised method that brings semantically-similar instances close while pushing dissimilar instances apart within a mini-batch. Its key to success is the negative sharing strategy, in which every instance serves as a negative for the others within the mini-batch. Recent studies aim to improve performance by sampling hard negatives \textit{within the current mini-batch}, whose quality is bounded by the mini-batch itself. In this work, we propose to improve contrastive learning by sampling mini-batches from the input data. We present BatchSampler\footnote{The code is available at \url{https://github.com/THUDM/BatchSampler}} to sample mini-batches of hard-to-distinguish (i.e., hard and true negatives to each other) instances. To make each mini-batch have fewer false negatives, we design the proximity graph of randomly-selected instances. To form the mini-batch, we leverage random walk with restart on the proximity graph to help sample hard-to-distinguish instances. BatchSampler is a simple and general technique that can be directly plugged into existing contrastive learning models in vision, language, and graphs. Extensive experiments on datasets of three modalities show that BatchSampler can consistently improve the performance of powerful contrastive models, as shown by significant improvements of SimCLR on ImageNet-100, SimCSE on STS (language), and GraphCL and MVGRL on graph datasets.
comment: 17 pages, 16 figures
☆ Click: Controllable Text Generation with Sequence Likelihood Contrastive Learning ACL 2023
It has always been an important yet challenging problem to control language models to avoid generating texts with undesirable attributes, such as toxic language and unnatural repetition. We introduce Click for controllable text generation, which needs no modification to the model architecture and facilitates out-of-the-box use of trained models. It employs a contrastive loss on sequence likelihood, which fundamentally decreases the generation probability of negative samples (i.e., generations with undesirable attributes). It also adopts a novel likelihood ranking-based strategy to construct contrastive samples from model generations. On the tasks of language detoxification, sentiment steering, and repetition reduction, we show that Click outperforms strong baselines of controllable text generation and demonstrate the superiority of Click's sample construction strategy.
comment: Findings of ACL 2023
☆ Inference-Time Intervention: Eliciting Truthful Answers from a Language Model
We introduce Inference-Time Intervention (ITI), a technique designed to enhance the truthfulness of large language models (LLMs). ITI operates by shifting model activations during inference, following a set of directions across a limited number of attention heads. This intervention significantly improves the performance of LLaMA models on the TruthfulQA benchmark. On an instruction-finetuned LLaMA called Alpaca, ITI improves its truthfulness from 32.5% to 65.1%. We identify a tradeoff between truthfulness and helpfulness and demonstrate how to balance it by tuning the intervention strength. ITI is minimally invasive and computationally inexpensive. Moreover, the technique is data efficient: while approaches like RLHF require extensive annotations, ITI locates truthful directions using only few hundred examples. Our findings suggest that LLMs may have an internal representation of the likelihood of something being true, even as they produce falsehoods on the surface.
comment: code: https://github.com/likenneth/honest_llama
☆ An Empirical Analysis of Parameter-Efficient Methods for Debiasing Pre-Trained Language Models ACL 2023
The increasingly large size of modern pretrained language models not only makes them inherit more human-like biases from the training corpora, but also makes it computationally expensive to mitigate such biases. In this paper, we investigate recent parameter-efficient methods in combination with counterfactual data augmentation (CDA) for bias mitigation. We conduct extensive experiments with prefix tuning, prompt tuning, and adapter tuning on different language models and bias types to evaluate their debiasing performance and abilities to preserve the internal knowledge of a pre-trained model. We find that the parameter-efficient methods (i) are effective in mitigating gender bias, where adapter tuning is consistently the most effective one and prompt tuning is more suitable for GPT-2 than BERT, (ii) are less effective when it comes to racial and religious bias, which may be attributed to the limitations of CDA, and (iii) can perform similarly to or sometimes better than full fine-tuning with improved time and memory efficiency, as well as maintain the internal knowledge in BERT and GPT-2, evaluated via fact retrieval and downstream fine-tuning.
comment: accepted to ACL 2023
☆ Augmenting Reddit Posts to Determine Wellness Dimensions impacting Mental Health
Amid ongoing health crisis, there is a growing necessity to discern possible signs of Wellness Dimensions (WD) manifested in self-narrated text. As the distribution of WD on social media data is intrinsically imbalanced, we experiment the generative NLP models for data augmentation to enable further improvement in the pre-screening task of classifying WD. To this end, we propose a simple yet effective data augmentation approach through prompt-based Generative NLP models, and evaluate the ROUGE scores and syntactic/semantic similarity among existing interpretations and augmented data. Our approach with ChatGPT model surpasses all the other methods and achieves improvement over baselines such as Easy-Data Augmentation and Backtranslation. Introducing data augmentation to generate more training samples and balanced dataset, results in the improved F-score and the Matthew's Correlation Coefficient for upto 13.11% and 15.95%, respectively.
☆ Active Sparse Conversations for Improved Audio-Visual Embodied Navigation
Efficient navigation towards an audio-goal necessitates an embodied agent to not only possess the ability to use audio-visual cues effectively, but also be equipped to actively (but occasionally) seek human/oracle assistance without sacrificing autonomy, e.g., when it is uncertain of where to navigate towards locating a noisy or sporadic audio goal. To this end, we present CAVEN -- a conversational audio-visual embodied navigation agent that is capable of posing navigation questions to a human/oracle and processing the oracle responses; both in free-form natural language. At the core of CAVEN is a multimodal hierarchical reinforcement learning (RL) setup that is equipped with a high-level policy that is trained to choose from one of three low-level policies (at every step), namely: (i) to navigate using audio-visual cues, or (ii) to frame a question to the oracle and receive a short or detailed response, or (iii) ask generic questions (when unsure of what to ask) and receive instructions. Key to generating the agent's questions is our novel TrajectoryNet that forecasts the most likely next steps to the goal and a QuestionNet that uses these steps to produce a question. All the policies are learned end-to-end via the RL setup, with penalties to enforce sparsity in receiving navigation instructions from the oracle. To evaluate the performance of CAVEN, we present extensive experiments on the SoundSpaces framework for the task of semantic audio-visual navigation. Our results show that CAVEN achieves upto 12% gain in performance over competing methods, especially in localizing new sound sources, even in the presence of auditory distractions.
☆ An Analysis of Reader Engagement in Literary Fiction through Eye Tracking and Linguistic Features
Capturing readers' engagement in fiction is a challenging but important aspect of narrative understanding. In this study, we collected 23 readers' reactions to 2 short stories through eye tracking, sentence-level annotations, and an overall engagement scale survey. We analyzed the significance of various qualities of the text in predicting how engaging a reader is likely to find it. As enjoyment of fiction is highly contextual, we also investigated individual differences in our data. Furthering our understanding of what captivates readers in fiction will help better inform models used in creative narrative generation and collaborative writing tools.
comment: 9 pages, 4 figures
☆ Triggering Multi-Hop Reasoning for Question Answering in Language Models using Soft Prompts and Random Walks ACL 2023
Despite readily memorizing world knowledge about entities, pre-trained language models (LMs) struggle to compose together two or more facts to perform multi-hop reasoning in question-answering tasks. In this work, we propose techniques that improve upon this limitation by relying on random walks over structured knowledge graphs. Specifically, we use soft prompts to guide LMs to chain together their encoded knowledge by learning to map multi-hop questions to random walk paths that lead to the answer. Applying our methods on two T5 LMs shows substantial improvements over standard tuning approaches in answering questions that require 2-hop reasoning.
comment: Findings of ACL 2023
☆ Sentiment Analysis in Finance: From Transformers Back to eXplainable Lexicons (XLex)
Lexicon-based sentiment analysis (SA) in finance leverages specialized, manually annotated lexicons created by human experts to extract sentiment from financial texts. Although lexicon-based methods are simple to implement and fast to operate on textual data, they require considerable manual annotation efforts to create, maintain, and update the lexicons. These methods are also considered inferior to the deep learning-based approaches, such as transformer models, which have become dominant in various NLP tasks due to their remarkable performance. However, transformers require extensive data and computational resources for both training and testing. Additionally, they involve significant prediction times, making them unsuitable for real-time production environments or systems with limited processing capabilities. In this paper, we introduce a novel methodology named eXplainable Lexicons (XLex) that combines the advantages of both lexicon-based methods and transformer models. We propose an approach that utilizes transformers and SHapley Additive exPlanations (SHAP) for explainability to learn financial lexicons. Our study presents four main contributions. Firstly, we demonstrate that transformer-aided explainable lexicons can enhance the vocabulary coverage of the benchmark Loughran-McDonald (LM) lexicon, reducing the human involvement in annotating, maintaining, and updating the lexicons. Secondly, we show that the resulting lexicon outperforms the standard LM lexicon in SA of financial datasets. Thirdly, we illustrate that the lexicon-based approach is significantly more efficient in terms of model speed and size compared to transformers. Lastly, the XLex approach is inherently more interpretable than transformer models as lexicon models rely on predefined rules, allowing for better insights into the results of SA and making the XLex approach a viable tool for financial decision-making.
comment: Submitted to IEEE Access
☆ Toward More Accurate and Generalizable Evaluation Metrics for Task-Oriented Dialogs
Measurement of interaction quality is a critical task for the improvement of spoken dialog systems. Existing approaches to dialog quality estimation either focus on evaluating the quality of individual turns, or collect dialog-level quality measurements from end users immediately following an interaction. In contrast to these approaches, we introduce a new dialog-level annotation workflow called Dialog Quality Annotation (DQA). DQA expert annotators evaluate the quality of dialogs as a whole, and also label dialogs for attributes such as goal completion and user sentiment. In this contribution, we show that: (i) while dialog quality cannot be completely decomposed into dialog-level attributes, there is a strong relationship between some objective dialog attributes and judgments of dialog quality; (ii) for the task of dialog-level quality estimation, a supervised model trained on dialog-level annotations outperforms methods based purely on aggregating turn-level features; and (iii) the proposed evaluation model shows better domain generalization ability compared to the baselines. On the basis of these results, we argue that having high-quality human-annotated data is an important component of evaluating interaction quality for large industrial-scale voice assistant platforms.
☆ Büyük dil modellerinin Türkçe verisetleri ile eğitilmesi ve ince ayarlanması
Large language models have advanced enormously, gained vast attraction and are having a phase of intensed research. Some of the developed models and training datasets have been made open-accessible. Hence these may be further fine-tuned with some techniques to obtain specialized models for specific tasks. When it comes to Turkish language, open-access models do not provide satisfactory coverage. This is also observed over published datasets. In this work, we propose some ideas to mitigate this issue: creating large Turkish datasets, training LLMs with these and fine-tuning pre-trained models with Turkish inputs. We report our findings on Turkish-based trainings with the problems encountered along the way. We conclude with outcomes of these experiments and propose ideas for further works. -- B\"uy\"uk dil modelleri inan{\i}lmaz \"ol\c{c}\"ude geli\c{s}mekte, b\"uy\"uk ilgi toplayarak ve \"uzerlerinde yo\u{g}un ara\c{s}tirmalarin yapildi\u{g}i bir d\"onemdedirler. Geli\c{s}tirilen modeller ve e\u{g}itimde kullanilan verisetlerinden bazilari a\c{c}ik eri\c{s}imli olarak sunulmaktadir. B\"oylece ince ayarlama teknikleri uygulayarak \"ozelle\c{s}mi\c{s} g\"orevler i\c{c}in \c{c}ali\c{s}abilir modeller elde edilmektedir. T\"urk\c{c}e s\"oz konusu oldu\u{g}unda bu modellerinin kapsayicili\u{g}i yeterli d\"uzeyde de\u{g}ildir. Bu durum, yayimlanan verisetlerinde de g\"ozlemlenebilir. Bunu a\c{s}manin yollari T\"urk\c{c}e i\c{c}erikli b\"uy\"uk verisetlerinin olu\c{s}turulmasi, b\"uy\"uk dil modellerinin bunlarla e\u{g}itilmesi ve \"onceden e\u{g}itilmi\c{s} modellerin T\"urk\c{c}e girdilerle ince ayarlanmalari olabilir. Bu \c{c}ali\c{s}mada a\c{c}ik eri\c{s}imli dil modelleri ve verisetleri \"uzerinde durulmakta ve T\"urk\c{c}e temelli bazi deneyler, kar\c{s}ila\c{s}ilan sorunlar ve sonu\c{c}lar irdelenmektedir.
comment: 7 pages
☆ Revisiting Conversation Discourse for Dialogue Disentanglement
Dialogue disentanglement aims to detach the chronologically ordered utterances into several independent sessions. Conversation utterances are essentially organized and described by the underlying discourse, and thus dialogue disentanglement requires the full understanding and harnessing of the intrinsic discourse attribute. In this paper, we propose enhancing dialogue disentanglement by taking full advantage of the dialogue discourse characteristics. First of all, \textbf{in feature encoding stage}, we construct the heterogeneous graph representations to model the various dialogue-specific discourse structural features, including the static speaker-role structures (i.e., speaker-utterance and speaker-mentioning structure) and the dynamic contextual structures (i.e., the utterance-distance and partial-replying structure). We then develop a structure-aware framework to integrate the rich structural features for better modeling the conversational semantic context. Second, \textbf{in model learning stage}, we perform optimization with a hierarchical ranking loss mechanism, which groups dialogue utterances into different discourse levels and carries training covering pair-wise and session-wise levels hierarchically. Third, \textbf{in inference stage}, we devise an easy-first decoding algorithm, which performs utterance pairing under the easy-to-hard manner with a global context, breaking the constraint of traditional sequential decoding order. On two benchmark datasets, our overall system achieves new state-of-the-art performances on all evaluations. In-depth analyses further demonstrate the efficacy of each proposed idea and also reveal how our methods help advance the task. Our work has great potential to facilitate broader multi-party multi-thread dialogue applications.
comment: under review
☆ TKDP: Threefold Knowledge-enriched Deep Prompt Tuning for Few-shot Named Entity Recognition
Few-shot named entity recognition (NER) exploits limited annotated instances to identify named mentions. Effectively transferring the internal or external resources thus becomes the key to few-shot NER. While the existing prompt tuning methods have shown remarkable few-shot performances, they still fail to make full use of knowledge. In this work, we investigate the integration of rich knowledge to prompt tuning for stronger few-shot NER. We propose incorporating the deep prompt tuning framework with threefold knowledge (namely TKDP), including the internal 1) context knowledge and the external 2) label knowledge & 3) sememe knowledge. TKDP encodes the three feature sources and incorporates them into the soft prompt embeddings, which are further injected into an existing pre-trained language model to facilitate predictions. On five benchmark datasets, our knowledge-enriched model boosts by at most 11.53% F1 over the raw deep prompt method, and significantly outperforms 8 strong-performing baseline systems in 5-/10-/20-shot settings, showing great potential in few-shot NER. Our TKDP can be broadly adapted to other few-shot tasks without effort.
comment: under review
☆ ECQED: Emotion-Cause Quadruple Extraction in Dialogs
The existing emotion-cause pair extraction (ECPE) task, unfortunately, ignores extracting the emotion type and cause type, while these fine-grained meta-information can be practically useful in real-world applications, i.e., chat robots and empathic dialog generation. Also the current ECPE is limited to the scenario of single text piece, while neglecting the studies at dialog level that should have more realistic values. In this paper, we extend the ECPE task with a broader definition and scenario, presenting a new task, Emotion-Cause Quadruple Extraction in Dialogs (ECQED), which requires detecting emotion-cause utterance pairs and emotion and cause types. We present an ECQED model based on a structural and semantic heterogeneous graph as well as a parallel grid tagging scheme, which advances in effectively incorporating the dialog context structure, meanwhile solving the challenging overlapped quadruple issue. Via experiments we show that introducing the fine-grained emotion and cause features evidently helps better dialog generation. Also our proposed ECQED system shows exceptional superiority over baselines on both the emotion-cause quadruple or pair extraction tasks, meanwhile being highly efficient.
comment: work in progress
☆ Leveraging Explicit Procedural Instructions for Data-Efficient Action Prediction
Task-oriented dialogues often require agents to enact complex, multi-step procedures in order to meet user requests. While large language models have found success automating these dialogues in constrained environments, their widespread deployment is limited by the substantial quantities of task-specific data required for training. The following paper presents a data-efficient solution to constructing dialogue systems, leveraging explicit instructions derived from agent guidelines, such as company policies or customer service manuals. Our proposed Knowledge-Augmented Dialogue System (KADS) combines a large language model with a knowledge retrieval module that pulls documents outlining relevant procedures from a predefined set of policies, given a user-agent interaction. To train this system, we introduce a semi-supervised pre-training scheme that employs dialogue-document matching and action-oriented masked language modeling with partial parameter freezing. We evaluate the effectiveness of our approach on prominent task-oriented dialogue datasets, Action-Based Conversations Dataset and Schema-Guided Dialogue, for two dialogue tasks: action state tracking and workflow discovery. Our results demonstrate that procedural knowledge augmentation improves accuracy predicting in- and out-of-distribution actions while preserving high performance in settings with low or sparse data.
☆ Recognition of Handwritten Japanese Characters Using Ensemble of Convolutional Neural Networks
The Japanese writing system is complex, with three character types of Hiragana, Katakana, and Kanji. Kanji consists of thousands of unique characters, further adding to the complexity of character identification and literature understanding. Being able to translate handwritten Japanese characters into digital text is useful for data analysis, translation, learning and cultural preservation. In this study, a machine learning approach to analyzing and recognizing handwritten Japanese characters (Kanji) is proposed. The study used an ensemble of three convolutional neural networks (CNNs) for recognizing handwritten Kanji characters and utilized four datasets of MNIST, K-MNIST, Kuzushiji-49 (K49) and the top 150 represented classes in the Kuzushiji-Kanji (K-Kanji) dataset for its performance evaluation. The results indicate feasibility of using proposed CNN-ensemble architecture for recognizing handwritten characters, achieving 99.4%, 96.4%, 95.0% and 96.4% classification accuracy on MNIST, K-MNIS, K49, and K-Kanji datasets respectively.
☆ MISGENDERED: Limits of Large Language Models in Understanding Pronouns ACL 2023
Content Warning: This paper contains examples of misgendering and erasure that could be offensive and potentially triggering. Gender bias in language technologies has been widely studied, but research has mostly been restricted to a binary paradigm of gender. It is essential also to consider non-binary gender identities, as excluding them can cause further harm to an already marginalized group. In this paper, we comprehensively evaluate popular language models for their ability to correctly use English gender-neutral pronouns (e.g., singular they, them) and neo-pronouns (e.g., ze, xe, thon) that are used by individuals whose gender identity is not represented by binary pronouns. We introduce MISGENDERED, a framework for evaluating large language models' ability to correctly use preferred pronouns, consisting of (i) instances declaring an individual's pronoun, followed by a sentence with a missing pronoun, and (ii) an experimental setup for evaluating masked and auto-regressive language models using a unified method. When prompted out-of-the-box, language models perform poorly at correctly predicting neo-pronouns (averaging 7.6% accuracy) and gender-neutral pronouns (averaging 31.0% accuracy). This inability to generalize results from a lack of representation of non-binary pronouns in training data and memorized associations. Few-shot adaptation with explicit examples in the prompt improves the performance but plateaus at only 45.4% for neo-pronouns. We release the full dataset, code, and demo at https://tamannahossainkay.github.io/misgendered/
comment: Accepted at ACL 2023 as a long paper
☆ Turning large language models into cognitive models
Large language models are powerful systems that excel at many tasks, ranging from translation to mathematical reasoning. Yet, at the same time, these models often show unhuman-like characteristics. In the present paper, we address this gap and ask whether large language models can be turned into cognitive models. We find that -- after finetuning them on data from psychological experiments -- these models offer accurate representations of human behavior, even outperforming traditional cognitive models in two decision-making domains. In addition, we show that their representations contain the information necessary to model behavior on the level of individual subjects. Finally, we demonstrate that finetuning on multiple tasks enables large language models to predict human behavior in a previously unseen task. Taken together, these results suggest that large, pre-trained models can be adapted to become generalist cognitive models, thereby opening up new research directions that could transform cognitive psychology and the behavioral sciences as a whole.
♻ ☆ A Watermark for Large Language Models ICML 2023
Potential harms of large language models can be mitigated by watermarking model output, i.e., embedding signals into generated text that are invisible to humans but algorithmically detectable from a short span of tokens. We propose a watermarking framework for proprietary language models. The watermark can be embedded with negligible impact on text quality, and can be detected using an efficient open-source algorithm without access to the language model API or parameters. The watermark works by selecting a randomized set of "green" tokens before a word is generated, and then softly promoting use of green tokens during sampling. We propose a statistical test for detecting the watermark with interpretable p-values, and derive an information-theoretic framework for analyzing the sensitivity of the watermark. We test the watermark using a multi-billion parameter model from the Open Pretrained Transformer (OPT) family, and discuss robustness and security.
comment: 13 pages in the main body. Published at ICML 2023. Code is available at github.com/jwkirchenbauer/lm-watermarking
♻ ☆ AutoPEFT: Automatic Configuration Search for Parameter-Efficient Fine-Tuning
Large pretrained language models are widely used in downstream NLP tasks via task-specific fine-tuning, but such procedures can be costly. Recently, Parameter-Efficient Fine-Tuning (PEFT) methods have achieved strong task performance while updating a much smaller number of parameters compared to full model fine-tuning (FFT). However, it is non-trivial to make informed design choices on the PEFT configurations, such as their architecture, the number of tunable parameters, and even the layers in which the PEFT modules are inserted. Consequently, it is highly likely that the current, manually designed configurations are suboptimal in terms of their performance-efficiency trade-off. Inspired by advances in neural architecture search, we propose AutoPEFT for automatic PEFT configuration selection: we first design an expressive configuration search space with multiple representative PEFT modules as building blocks. Using multi-objective Bayesian optimisation in a low-cost setup, we then discover a Pareto-optimal set of configurations with strong performance-cost trade-offs across different numbers of parameters that are also highly transferable across different tasks. Empirically, on GLUE and SuperGLUE tasks, we show that AutoPEFT-discovered configurations significantly outperform existing PEFT methods and are on par or better than FFT, without incurring substantial training efficiency costs.
comment: 17 pages, 7 figures, 9 tables
♻ ☆ oBERTa: Improving Sparse Transfer Learning via improved initialization, distillation, and pruning regimes ACL 2023
In this paper, we introduce the range of oBERTa language models, an easy-to-use set of language models which allows Natural Language Processing (NLP) practitioners to obtain between 3.8 and 24.3 times faster models without expertise in model compression. Specifically, oBERTa extends existing work on pruning, knowledge distillation, and quantization and leverages frozen embeddings improves distillation and model initialization to deliver higher accuracy on a broad range of transfer tasks. In generating oBERTa, we explore how the highly optimized RoBERTa differs from the BERT for pruning during pre-training and finetuning. We find it less amenable to compression during fine-tuning. We explore the use of oBERTa on seven representative NLP tasks and find that the improved compression techniques allow a pruned oBERTa model to match the performance of BERTbase and exceed the performance of Prune OFA Large on the SQUAD V1.1 Question Answering dataset, despite being 8x and 2x, respectively faster in inference. We release our code, training regimes, and associated model for broad usage to encourage usage and experimentation
comment: SustaiNLP2023 @ ACL 2023,9 pages, 2 figures, 45 tables
♻ ☆ Enhancing Programming eTextbooks with ChatGPT Generated Counterfactual-Thinking-Inspired Questions
Digital textbooks have become an integral part of everyday learning tasks. In this work, we consider the use of digital textbooks for programming classes. Generally, students struggle with utilizing textbooks on programming to the maximum, with a possible reason being that the example programs provided as illustration of concepts in these textbooks don't offer sufficient interactivity for students, and thereby not sufficiently motivating to explore or understand these programming examples better. In our work, we explore the idea of enhancing the navigability of intelligent textbooks with the use of ``counterfactual'' questions, to make students think critically about these programs and enhance possible program comprehension. Inspired from previous works on nudging students on counter factual thinking, we present the possibility to enhance digital textbooks with questions generated using GPT.
comment: Paper Under Review
♻ ☆ Detecting Harmful Content On Online Platforms: What Platforms Need Vs. Where Research Efforts Go
The proliferation of harmful content on online platforms is a major societal problem, which comes in many different forms including hate speech, offensive language, bullying and harassment, misinformation, spam, violence, graphic content, sexual abuse, self harm, and many other. Online platforms seek to moderate such content to limit societal harm, to comply with legislation, and to create a more inclusive environment for their users. Researchers have developed different methods for automatically detecting harmful content, often focusing on specific sub-problems or on narrow communities, as what is considered harmful often depends on the platform and on the context. We argue that there is currently a dichotomy between what types of harmful content online platforms seek to curb, and what research efforts there are to automatically detect such content. We thus survey existing methods as well as content moderation policies by online platforms in this light and we suggest directions for future work.
comment: The paper has been accepted for publication to ACM Computing Surveys (CSUR)
♻ ☆ Make Your Pre-trained Model Reversible: From Parameter to Memory Efficient Fine-Tuning
Parameter-efficient fine-tuning (PEFT) of pre-trained language models (PLMs) has emerged as a highly successful approach, with training only a small number of parameters without sacrificing performance and becoming the de-facto learning paradigm with the increasing size of PLMs. However, existing PEFT methods are not memory-efficient, because they still require caching most of the intermediate activations for the gradient calculation, akin to fine-tuning. One effective way to reduce the activation memory is to apply a reversible model, so the intermediate activations are not necessary to be cached and can be recomputed. Nevertheless, modifying a PLM to its reversible variant with PEFT is not straightforward, since the reversible model has a distinct architecture from the currently released PLMs. In this paper, we first investigate what is a key factor for the success of existing PEFT methods, and realize that it's essential to preserve the PLM's starting point when initializing a PEFT method. With this finding, we propose memory-efficient fine-tuning (MEFT) that inserts adapters into a PLM, preserving the PLM's starting point and making it reversible without additional pre-training. We evaluate MEFT on the GLUE benchmark and five question-answering tasks with various backbones, BERT, RoBERTa, BART and OPT. MEFT significantly reduces the activation memory up to 84% of full fine-tuning with a negligible amount of trainable parameters. Moreover, MEFT achieves the same score on GLUE and a comparable score on the question-answering tasks as full fine-tuning.
comment: Code at https://github.com/BaohaoLiao/mefts
♻ ☆ Explanation-based Finetuning Makes Models More Robust to Spurious Cues
Large Language Models (LLMs) are so powerful that they sometimes learn correlations between labels and features that are irrelevant to the task, leading to poor generalization on out-of-distribution data. We propose explanation-based finetuning as a general approach to mitigate LLMs' reliance on spurious correlations. Unlike standard finetuning where the model only predicts the answer given the input, we finetune the model to additionally generate a free-text explanation supporting its answer. To evaluate our method, we finetune the model on artificially constructed training sets containing different types of spurious cues, and test it on a test set without these cues. Compared to standard finetuning, our method makes GPT-3 (davinci) remarkably more robust against spurious cues in terms of accuracy drop across four classification tasks: ComVE (+1.2), CREAK (+9.1), e-SNLI (+15.4), and SBIC (+6.5). The efficacy generalizes across multiple model families and scales, with greater gains for larger models. Finally, our method also works well with explanations generated by the model, implying its applicability to more datasets without human-written explanations.
♻ ☆ Chain-of-Symbol Prompting Elicits Planning in Large Langauge Models
In this paper, we take the initiative to investigate the performance of LLMs on complex planning tasks that require LLMs to understand a virtual spatial environment simulated via natural language and act correspondingly in text. We propose a benchmark named Natural Language Planning and Action (Natala) composed of a set of novel tasks: Brick World, NLVR-based Manipulations, and Natural Language Navigation. We found that current popular LLMs such as ChatGPT still lack abilities in complex planning. This arises a question -- do the LLMs have a good understanding of the environments described in natural language, or maybe other alternatives such as symbolic representations are neater and hence better to be understood by LLMs? To this end, we propose a novel method called CoS (Chain-of-Symbol Prompting) that represents the complex environments with condensed symbolic spatial representations during the chained intermediate thinking steps. CoS is easy to use and does not need additional training on LLMs. Extensive experiments indicate that CoS clearly surpasses the performance of the Chain-of-Thought (CoT) Prompting in all three planning tasks with even fewer tokens used in the inputs compared with CoT on ChatGPT and InstructGPT. The performance gain is strong, by up to 60.8% accuracy (from 31.8% to 92.6%) on Brick World for ChatGPT. CoS also reduces the number of tokens in the prompt obviously, by up to 65.8% of the tokens (from 407 to 139) for the intermediate steps from demonstrations on Brick World.
♻ ☆ Self-Adaptive Named Entity Recognition by Retrieving Unstructured Knowledge EACL2023
Although named entity recognition (NER) helps us to extract domain-specific entities from text (e.g., artists in the music domain), it is costly to create a large amount of training data or a structured knowledge base to perform accurate NER in the target domain. Here, we propose self-adaptive NER, which retrieves external knowledge from unstructured text to learn the usages of entities that have not been learned well. To retrieve useful knowledge for NER, we design an effective two-stage model that retrieves unstructured knowledge using uncertain entities as queries. Our model predicts the entities in the input and then finds those of which the prediction is not confident. Then, it retrieves knowledge by using these uncertain entities as queries and concatenates the retrieved text to the original input to revise the prediction. Experiments on CrossNER datasets demonstrated that our model outperforms strong baselines by 2.35 points in F1 metric.
comment: EACL2023 (long)
♻ ☆ DP-BART for Privatized Text Rewriting under Local Differential Privacy ACL
Privatized text rewriting with local differential privacy (LDP) is a recent approach that enables sharing of sensitive textual documents while formally guaranteeing privacy protection to individuals. However, existing systems face several issues, such as formal mathematical flaws, unrealistic privacy guarantees, privatization of only individual words, as well as a lack of transparency and reproducibility. In this paper, we propose a new system 'DP-BART' that largely outperforms existing LDP systems. Our approach uses a novel clipping method, iterative pruning, and further training of internal representations which drastically reduces the amount of noise required for DP guarantees. We run experiments on five textual datasets of varying sizes, rewriting them at different privacy guarantees and evaluating the rewritten texts on downstream text classification tasks. Finally, we thoroughly discuss the privatized text rewriting approach and its limitations, including the problem of the strict text adjacency constraint in the LDP paradigm that leads to the high noise requirement.
comment: Accepted at ACL Findings 2023
♻ ☆ MERT: Acoustic Music Understanding Model with Large-Scale Self-supervised Training
Self-supervised learning (SSL) has recently emerged as a promising paradigm for training generalisable models on large-scale data in the fields of vision, text, and speech. Although SSL has been proven effective in speech and audio, its application to music audio has yet to be thoroughly explored. This is primarily due to the distinctive challenges associated with modelling musical knowledge, particularly its tonal and pitched characteristics of music. To address this research gap, we propose an acoustic Music undERstanding model with large-scale self-supervised Training (MERT), which incorporates teacher models to provide pseudo labels in the masked language modelling (MLM) style acoustic pre-training. In our exploration, we identified a superior combination of teacher models, which outperforms conventional speech and audio approaches in terms of performance. This combination includes an acoustic teacher based on Residual Vector Quantization - Variational AutoEncoder (RVQ-VAE) and a musical teacher based on the Constant-Q Transform (CQT). These teachers effectively guide our student model, a BERT-style transformer encoder, to better model music audio. In addition, we introduce an in-batch noise mixture augmentation to enhance the representation robustness. Furthermore, we explore a wide range of settings to overcome the instability in acoustic language model pre-training, which allows our designed paradigm to scale from 95M to 330M parameters. Experimental results indicate that our model can generalise and perform well on 14 music understanding tasks and attains state-of-the-art (SOTA) overall scores. The code and models are online: https://github.com/yizhilll/MERT.
♻ ☆ How poor is the stimulus? Evaluating hierarchical generalization in neural networks trained on child-directed speech ACL
When acquiring syntax, children consistently choose hierarchical rules over competing non-hierarchical possibilities. Is this preference due to a learning bias for hierarchical structure, or due to more general biases that interact with hierarchical cues in children's linguistic input? We explore these possibilities by training LSTMs and Transformers - two types of neural networks without a hierarchical bias - on data similar in quantity and content to children's linguistic input: text from the CHILDES corpus. We then evaluate what these models have learned about English yes/no questions, a phenomenon for which hierarchical structure is crucial. We find that, though they perform well at capturing the surface statistics of child-directed speech (as measured by perplexity), both model types generalize in a way more consistent with an incorrect linear rule than the correct hierarchical rule. These results suggest that human-like generalization from text alone requires stronger biases than the general sequence-processing biases of standard neural network architectures.
comment: 10 pages plus references and appendices; accepted to ACL
♻ ☆ Aligning Language Models with Preferences through f-divergence Minimization
Aligning language models with preferences can be posed as approximating a target distribution representing some desired behavior. Existing approaches differ both in the functional form of the target distribution and the algorithm used to approximate it. For instance, Reinforcement Learning from Human Feedback (RLHF) corresponds to minimizing a reverse KL from an implicit target distribution arising from a KL penalty in the objective. On the other hand, Generative Distributional Control (GDC) has an explicit target distribution and minimizes a forward KL from it using the Distributional Policy Gradient (DPG) algorithm. In this paper, we propose a new approach, f-DPG, which allows the use of any f-divergence to approximate any target distribution that can be evaluated. f-DPG unifies both frameworks (RLHF, GDC) and the approximation methods (DPG, RL with KL penalties). We show the practical benefits of various choices of divergence objectives and demonstrate that there is no universally optimal objective but that different divergences present different alignment and diversity trade-offs. We show that Jensen-Shannon divergence strikes a good balance between these objectives, and frequently outperforms forward KL divergence by a wide margin, leading to significant improvements over prior work. These distinguishing characteristics between divergences persist as the model size increases, highlighting the importance of selecting appropriate divergence objectives.
♻ ☆ GigaST: A 10,000-hour Pseudo Speech Translation Corpus
This paper introduces GigaST, a large-scale pseudo speech translation (ST) corpus. We create the corpus by translating the text in GigaSpeech, an English ASR corpus, into German and Chinese. The training set is translated by a strong machine translation system and the test set is translated by human. ST models trained with an addition of our corpus obtain new state-of-the-art results on the MuST-C English-German benchmark test set. We provide a detailed description of the translation process and verify its quality. We make the translated text data public and hope to facilitate research in speech translation. Additionally, we also release the training scripts on NeurST to make it easy to replicate our systems. GigaST dataset is available at https://st-benchmark.github.io/resources/GigaST.
comment: Accepted at Interspeech 2023. GigaST dataset is available at https://st-benchmark.github.io/resources/GigaST
♻ ☆ A Comprehensive Survey on Deep Learning for Relation Extraction: Recent Advances and New Frontiers
Relation extraction (RE) involves identifying the relations between entities from unstructured texts. RE serves as the foundation for many natural language processing (NLP) applications, such as knowledge graph completion, question answering, and information retrieval. In recent years, deep neural networks have dominated the field of RE and made noticeable progress. Subsequently, the large pre-trained language models (PLMs) have taken the state-of-the-art of RE to a new level. This survey provides a comprehensive review of existing deep learning techniques for RE. First, we introduce RE resources, including RE datasets and evaluation metrics. Second, we propose a new taxonomy to categorize existing works from three perspectives (text representation, context encoding, and triplet prediction). Third, we discuss several important challenges faced by RE and summarize potential techniques to tackle these challenges. Finally, we outline some promising future directions and prospects in this field. This survey is expected to facilitate researchers' collaborative efforts to tackle the challenges of real-life RE systems.
♻ ☆ Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding
We present Video-LLaMA, a multi-modal framework that empowers Large Language Models (LLMs) with the capability of understanding both visual and auditory content in the video. Video-LLaMA bootstraps cross-modal training from the frozen pre-trained visual & audio encoders and the frozen LLMs. Unlike previous vision-LLMs that focus on static image comprehensions such as MiniGPT-4 and LLaVA, Video-LLaMA mainly tackles two challenges in video understanding: (1) capturing the temporal changes in visual scenes, (2) integrating audio-visual signals. To counter the first challenge, we propose a Video Q-former to assemble the pre-trained image encoder into our video encoder and introduce a video-to-text generation task to learn video-language correspondence. For the second challenge, we leverage ImageBind, a universal embedding model aligning multiple modalities as the pre-trained audio encoder, and introduce an Audio Q-former on top of ImageBind to learn reasonable auditory query embeddings for the LLM module. To align the output of both visual & audio encoders with LLM's embedding space, we train Video-LLaMA on massive video/image-caption pairs as well as visual-instruction-tuning datasets of moderate amount but higher quality. We found Video-LLaMA showcases the ability to perceive and comprehend video content, generating meaningful responses that are grounded in the visual and auditory information presented in the videos. This highlights the potential of Video-LLaMA as a promising prototype for audio-visual AI assistants.
comment: Technical Report; Code, Pretrained Model, and Dataset: https://github.com/DAMO-NLP-SG/Video-LLaMA
♻ ☆ SI-LSTM: Speaker Hybrid Long-short Term Memory and Cross Modal Attention for Emotion Recognition in Conversation
Emotion Recognition in Conversation~(ERC) across modalities is of vital importance for a variety of applications, including intelligent healthcare, artificial intelligence for conversation, and opinion mining over chat history. The crux of ERC is to model both cross-modality and cross-time interactions throughout the conversation. Previous methods have made progress in learning the time series information of conversation while lacking the ability to trace down the different emotional states of each speaker in a conversation. In this paper, we propose a recurrent structure called Speaker Information Enhanced Long-Short Term Memory (SI-LSTM) for the ERC task, where the emotional states of the distinct speaker can be tracked in a sequential way to enhance the learning of the emotion in conversation. Further, to improve the learning of multimodal features in ERC, we utilize a cross-modal attention component to fuse the features between different modalities and model the interaction of the important information from different modalities. Experimental results on two benchmark datasets demonstrate the superiority of the proposed SI-LSTM against the state-of-the-art baseline methods in the ERC task on multimodal data.
comment: modification needed
♻ ☆ Can In-context Learners Learn a Reasoning Concept from Demonstrations? ACL 2023
Large language models show an emergent ability to learn a new task from a small number of input-output demonstrations. However, recent work shows that in-context learners largely rely on their pre-trained knowledge, such as the sentiment of the labels, instead of finding new associations in the input. However, the commonly-used few-shot evaluation settings using a random selection of in-context demonstrations can not disentangle models' ability to learn a new skill from demonstrations, as most of the randomly-selected demonstrations do not present relations informative for prediction beyond exposing the new task distribution. To disentangle models' in-context learning ability independent of models' memory, we introduce a Conceptual few-shot learning method selecting the demonstrations sharing a possibly-informative concept with the predicted sample. We extract a set of such concepts from annotated explanations and measure how much can models benefit from presenting these concepts in few-shot demonstrations. We find that smaller models are more sensitive to the presented concepts. While some of the models are able to benefit from concept-presenting demonstrations for each assessed concept, we find that none of the assessed in-context learners can benefit from all presented reasoning concepts consistently, leaving the in-context concept learning an open challenge.
comment: Accepted at ACL 2023 Natural Language Reasoning workshop
♻ ☆ Graph2topic: an opensource topic modeling framework based on sentence embedding and community detection
It has been reported that clustering-based topic models, which cluster high-quality sentence embeddings with an appropriate word selection method, can generate better topics than generative probabilistic topic models. However, these approaches suffer from the inability to select appropriate parameters and incomplete models that overlook the quantitative relation between words with topics and topics with text. To solve these issues, we propose graph to topic (G2T), a simple but effective framework for topic modelling. The framework is composed of four modules. First, document representation is acquired using pretrained language models. Second, a semantic graph is constructed according to the similarity between document representations. Third, communities in document semantic graphs are identified, and the relationship between topics and documents is quantified accordingly. Fourth, the word--topic distribution is computed based on a variant of TFIDF. Automatic evaluation suggests that G2T achieved state-of-the-art performance on both English and Chinese documents with different lengths.
comment: 11pages
♻ ☆ The Politics of Language Choice: How the Russian-Ukrainian War Influences Ukrainians' Language Use on Twitter
The use of language is innately political and often a vehicle of cultural identity as well as the basis for nation building. Here, we examine language choice and tweeting activity of Ukrainian citizens based on more than 4 million geo-tagged tweets from over 62,000 users before and during the Russian-Ukrainian War, from January 2020 to October 2022. Using statistical models, we disentangle sample effects, arising from the in- and outflux of users on Twitter, from behavioural effects, arising from behavioural changes of the users. We observe a steady shift from the Russian language towards the Ukrainian language already before the war, which drastically speeds up with its outbreak. We attribute these shifts in large part to users' behavioural changes. Notably, we find that more than half of the Russian-tweeting users shift towards Ukrainian as a result of the war.
♻ ☆ Topological Data Analysis for Speech Processing INTERSPEECH 2023
We apply topological data analysis (TDA) to speech classification problems and to the introspection of a pretrained speech model, HuBERT. To this end, we introduce a number of topological and algebraic features derived from Transformer attention maps and embeddings. We show that a simple linear classifier built on top of such features outperforms a fine-tuned classification head. In particular, we achieve an improvement of about $9\%$ accuracy and $5\%$ ERR on four common datasets; on CREMA-D, the proposed feature set reaches a new state of the art performance with accuracy $80.155$. We also show that topological features are able to reveal functional roles of speech Transformer heads; e.g., we find the heads capable to distinguish between pairs of sample sources (natural/synthetic) or voices without any downstream fine-tuning. Our results demonstrate that TDA is a promising new approach for speech analysis, especially for tasks that require structural prediction. Appendices, an introduction to TDA, and other additional materials are available here - https://topohubert.github.io/speech-topology-webpages/
comment: Accepted to INTERSPEECH 2023 conference
♻ ☆ Guess the Instruction! Flipped Learning Makes Language Models Stronger Zero-Shot Learners ICLR 2023
Meta-training, which fine-tunes the language model (LM) on various downstream tasks by maximizing the likelihood of the target label given the task instruction and input instance, has improved the zero-shot task generalization performance. However, meta-trained LMs still struggle to generalize to challenging tasks containing novel labels unseen during meta-training. In this paper, we propose Flipped Learning, an alternative method of meta-training which trains the LM to generate the task instruction given the input instance and label. During inference, the LM trained with Flipped Learning, referred to as Flipped, selects the label option that is most likely to generate the task instruction. On 14 tasks of the BIG-bench benchmark, the 11B-sized Flipped outperforms zero-shot T0-11B and even a 16 times larger 3-shot GPT-3 (175B) on average by 8.4% and 9.7% points, respectively. Flipped gives particularly large improvements on tasks with unseen labels, outperforming T0-11B by up to +20% average F1 score. This indicates that the strong task generalization of Flipped comes from improved generalization to novel labels. We release our code at https://github.com/seonghyeonye/Flipped-Learning.
comment: ICLR 2023
♻ ☆ Self-supervised Predictive Coding Models Encode Speaker and Phonetic Information in Orthogonal Subspaces
Self-supervised speech representations are known to encode both speaker and phonetic information, but how they are distributed in the high-dimensional space remains largely unexplored. We hypothesize that they are encoded in orthogonal subspaces, a property that lends itself to simple disentanglement. Applying principal component analysis to representations of two predictive coding models, we identify two subspaces that capture speaker and phonetic variances, and confirm that they are nearly orthogonal. Based on this property, we propose a new speaker normalization method which collapses the subspace that encodes speaker information, without requiring transcriptions. Probing experiments show that our method effectively eliminates speaker information and outperforms a previous baseline in phone discrimination tasks. Moreover, the approach generalizes and can be used to remove information of unseen speakers.
comment: Accepted to Interspeech 2023
♻ ☆ A Kernel-Based View of Language Model Fine-Tuning ICML 2023
It has become standard to solve NLP tasks by fine-tuning pre-trained language models (LMs), especially in low-data settings. There is minimal theoretical understanding of empirical success, e.g., why fine-tuning a model with $10^8$ or more parameters on a couple dozen training points does not result in overfitting. We investigate whether the Neural Tangent Kernel (NTK) - which originated as a model to study the gradient descent dynamics of infinitely wide networks with suitable random initialization - describes fine-tuning of pre-trained LMs. This study was inspired by the decent performance of NTK for computer vision tasks (Wei et al., 2022). We extend the NTK formalism to Adam and use Tensor Programs (Yang, 2020) to characterize conditions under which the NTK lens may describe fine-tuning updates to pre-trained language models. Extensive experiments on 14 NLP tasks validate our theory and show that formulating the downstream task as a masked word prediction problem through prompting often induces kernel-based dynamics during fine-tuning. Finally, we use this kernel view to propose an explanation for the success of parameter-efficient subspace-based fine-tuning methods.
comment: Accepted at ICML 2023. Code and pre-computed kernels are publicly available at https://github.com/princeton-nlp/LM-Kernel-FT
♻ ☆ Large Language Models Can Be Easily Distracted by Irrelevant Context ICML 2023
Large language models have achieved impressive performance on various natural language processing tasks. However, so far they have been evaluated primarily on benchmarks where all information in the input context is relevant for solving the task. In this work, we investigate the distractibility of large language models, i.e., how the model problem-solving accuracy can be influenced by irrelevant context. In particular, we introduce Grade-School Math with Irrelevant Context (GSM-IC), an arithmetic reasoning dataset with irrelevant information in the problem description. We use this benchmark to measure the distractibility of cutting-edge prompting techniques for large language models, and find that the model performance is dramatically decreased when irrelevant information is included. We also identify several approaches for mitigating this deficiency, such as decoding with self-consistency and adding to the prompt an instruction that tells the language model to ignore the irrelevant information.
comment: Published in ICML 2023
♻ ☆ Benchmarking Spatial Relationships in Text-to-Image Generation
Spatial understanding is a fundamental aspect of computer vision and integral for human-level reasoning about images, making it an important component for grounded language understanding. While recent text-to-image synthesis (T2I) models have shown unprecedented improvements in photorealism, it is unclear whether they have reliable spatial understanding capabilities. We investigate the ability of T2I models to generate correct spatial relationships among objects and present VISOR, an evaluation metric that captures how accurately the spatial relationship described in text is generated in the image. To benchmark existing models, we introduce a dataset, SR2D, that contains sentences describing two objects and the spatial relationship between them. We construct an automated evaluation pipeline to recognize objects and their spatial relationships, and employ it in a large-scale evaluation of T2I models. Our experiments reveal a surprising finding that, although state-of-the-art T2I models exhibit high image quality, they are severely limited in their ability to generate multiple objects or the specified spatial relations between them. Our analyses demonstrate several biases and artifacts of T2I models such as the difficulty with generating multiple objects, a bias towards generating the first object mentioned, spatially inconsistent outputs for equivalent relationships, and a correlation between object co-occurrence and spatial understanding capabilities. We conduct a human study that shows the alignment between VISOR and human judgement about spatial understanding. We offer the SR2D dataset and the VISOR metric to the community in support of T2I reasoning research.
comment: preprint; Code and Data at https://github.com/microsoft/VISOR and https://huggingface.co/datasets/tgokhale/sr2d_visor
♻ ☆ Revisiting the Gold Standard: Grounding Summarization Evaluation with Robust Human Evaluation ACL 2023
Human evaluation is the foundation upon which the evaluation of both summarization systems and automatic metrics rests. However, existing human evaluation studies for summarization either exhibit a low inter-annotator agreement or have insufficient scale, and an in-depth analysis of human evaluation is lacking. Therefore, we address the shortcomings of existing summarization evaluation along the following axes: (1) We propose a modified summarization salience protocol, Atomic Content Units (ACUs), which is based on fine-grained semantic units and allows for a high inter-annotator agreement. (2) We curate the Robust Summarization Evaluation (RoSE) benchmark, a large human evaluation dataset consisting of 22,000 summary-level annotations over 28 top-performing systems on three datasets. (3) We conduct a comparative study of four human evaluation protocols, underscoring potential confounding factors in evaluation setups. (4) We evaluate 50 automatic metrics and their variants using the collected human annotations across evaluation protocols and demonstrate how our benchmark leads to more statistically stable and significant results. The metrics we benchmarked include recent methods based on large language models (LLMs), GPTScore and G-Eval. Furthermore, our findings have important implications for evaluating LLMs, as we show that LLMs adjusted by human feedback (e.g., GPT-3.5) may overfit unconstrained human evaluation, which is affected by the annotators' prior, input-agnostic preferences, calling for more robust, targeted evaluation methods.
comment: ACL 2023 Camera Ready
♻ ☆ CCpdf: Building a High Quality Corpus for Visually Rich Documents from Web Crawl Data ICDAR 2023
In recent years, the field of document understanding has progressed a lot. A significant part of this progress has been possible thanks to the use of language models pretrained on large amounts of documents. However, pretraining corpora used in the domain of document understanding are single domain, monolingual, or nonpublic. Our goal in this paper is to propose an efficient pipeline for creating a big-scale, diverse, multilingual corpus of PDF files from all over the Internet using Common Crawl, as PDF files are the most canonical types of documents as considered in document understanding. We analysed extensively all of the steps of the pipeline and proposed a solution which is a trade-off between data quality and processing time. We also share a CCpdf corpus in a form or an index of PDF files along with a script for downloading them, which produces a collection useful for language model pretraining. The dataset and tools published with this paper offer researchers the opportunity to develop even better multilingual language models.
comment: Accepted at ICDAR 2023
♻ ☆ MultiLegalPile: A 689GB Multilingual Legal Corpus
Large, high-quality datasets are crucial for training Large Language Models (LLMs). However, so far, there are few datasets available for specialized critical domains such as law and the available ones are often only for the English language. We curate and release MultiLegalPile, a 689GB corpus in 24 languages from 17 jurisdictions. The MultiLegalPile corpus, which includes diverse legal data sources with varying licenses, allows for pretraining NLP models under fair use, with more permissive licenses for the Eurlex Resources and Legal mC4 subsets. We pretrain two RoBERTa models and one Longformer multilingually, and 24 monolingual models on each of the language-specific subsets and evaluate them on LEXTREME. Additionally, we evaluate the English and multilingual models on LexGLUE. Our multilingual models set a new SotA on LEXTREME and our English models on LexGLUE. We release the dataset, the trained models, and all of the code under the most open possible licenses.
♻ ☆ NAIST-SIC-Aligned: Automatically-Aligned English-Japanese Simultaneous Interpretation Corpus
It remains a question that how simultaneous interpretation (SI) data affects simultaneous machine translation (SiMT). Research has been limited due to the lack of a large-scale training corpus. In this work, we aim to fill in the gap by introducing NAIST-SIC-Aligned, which is an automatically-aligned parallel English-Japanese SI dataset. Starting with a non-aligned corpus NAIST-SIC, we propose a two-stage alignment approach to make the corpus parallel and thus suitable for model training. The first stage is coarse alignment where we perform a many-to-many mapping between source and target sentences, and the second stage is fine-grained alignment where we perform intra- and inter-sentence filtering to improve the quality of aligned pairs. To ensure the quality of the corpus, each step has been validated either quantitatively or qualitatively. This is the first open-sourced large-scale parallel SI dataset in the literature. We also manually curated a small test set for evaluation purposes. We hope our work advances research on SI corpora construction and SiMT. Please find our data at \url{https://github.com/mingzi151/AHC-SI}.
comment: Fixed typos
♻ ☆ DuNST: Dual Noisy Self Training for Semi-Supervised Controllable Text Generation
Self-training (ST) has prospered again in language understanding by augmenting the fine-tuning of pre-trained language models when labeled data is insufficient. However, it remains challenging to incorporate ST into attribute-controllable language generation. Augmented by only self-generated pseudo text, generation models over-emphasize exploitation of the previously learned space, suffering from a constrained generalization boundary. We revisit ST and propose a novel method, DuNST to alleviate this problem. DuNST jointly models text generation and classification with a shared Variational AutoEncoder and corrupts the generated pseudo text by two kinds of flexible noise to disturb the space. In this way, our model could construct and utilize both pseudo text from given labels and pseudo labels from available unlabeled text, which are gradually refined during the ST process. We theoretically demonstrate that DuNST can be regarded as enhancing exploration towards the potential real text space, providing a guarantee of improved performance. Experiments on three controllable generation tasks show that DuNST could significantly boost control accuracy while maintaining comparable generation fluency and diversity against several strong baselines.
♻ ☆ Learning Multi-Step Reasoning by Solving Arithmetic Tasks ACL 2023
Mathematical reasoning is regarded as a necessary ability for Language Models (LMs). Recent works demonstrate large LMs' impressive performance in solving math problems. The success is attributed to their Chain-of-Thought (CoT) reasoning abilities, i.e., the ability to decompose complex questions into step-by-step reasoning chains, but such ability seems only to emerge from models with abundant parameters. This work investigates how to incorporate relatively small LMs with the capabilities of multi-step reasoning. We propose to inject such abilities by continually pre-training LMs on a synthetic dataset MsAT which is composed of Multi-step Arithmetic Tasks. Our experiments on four math word problem datasets show the effectiveness of the proposed method in enhancing LMs' math reasoning abilities.
comment: ACL 2023. Code and data are available at https://github.com/TianduoWang/MsAT
♻ ☆ CapText: Large Language Model-based Caption Generation From Image Context and Description
While deep-learning models have been shown to perform well on image-to-text datasets, it is difficult to use them in practice for captioning images. This is because captions traditionally tend to be context-dependent and offer complementary information about an image, while models tend to produce descriptions that describe the visual features of the image. Prior research in caption generation has explored the use of models that generate captions when provided with the images alongside their respective descriptions or contexts. We propose and evaluate a new approach, which leverages existing large language models to generate captions from textual descriptions and context alone, without ever processing the image directly. We demonstrate that after fine-tuning, our approach outperforms current state-of-the-art image-text alignment models like OSCAR-VinVL on this task on the CIDEr metric.
comment: Update 6/6/23: Fixed typographic error in abstract
♻ ☆ A Technical Report for Polyglot-Ko: Open-Source Large-Scale Korean Language Models
Polyglot is a pioneering project aimed at enhancing the non-English language performance of multilingual language models. Despite the availability of various multilingual models such as mBERT (Devlin et al., 2019), XGLM (Lin et al., 2022), and BLOOM (Scao et al., 2022), researchers and developers often resort to building monolingual models in their respective languages due to the dissatisfaction with the current multilingual models non-English language capabilities. Addressing this gap, we seek to develop advanced multilingual language models that offer improved performance in non-English languages. In this paper, we introduce the Polyglot Korean models, which represent a specific focus rather than being multilingual in nature. In collaboration with TUNiB, our team collected 1.2TB of Korean data meticulously curated for our research journey. We made a deliberate decision to prioritize the development of Korean models before venturing into multilingual models. This choice was motivated by multiple factors: firstly, the Korean models facilitated performance comparisons with existing multilingual models; and finally, they catered to the specific needs of Korean companies and researchers. This paper presents our work in developing the Polyglot Korean models, which propose some steps towards addressing the non-English language performance gap in multilingual language models.
♻ ☆ Do GPTs Produce Less Literal Translations? ACL 2023
Large Language Models (LLMs) such as GPT-3 have emerged as general-purpose language models capable of addressing many natural language generation or understanding tasks. On the task of Machine Translation (MT), multiple works have investigated few-shot prompting mechanisms to elicit better translations from LLMs. However, there has been relatively little investigation on how such translations differ qualitatively from the translations generated by standard Neural Machine Translation (NMT) models. In this work, we investigate these differences in terms of the literalness of translations produced by the two systems. Using literalness measures involving word alignment and monotonicity, we find that translations out of English (E-X) from GPTs tend to be less literal, while exhibiting similar or better scores on MT quality metrics. We demonstrate that this finding is borne out in human evaluations as well. We then show that these differences are especially pronounced when translating sentences that contain idiomatic expressions.
comment: ACL 2023
♻ ☆ A Universal Discriminator for Zero-Shot Generalization ACL 2023
Generative modeling has been the dominant approach for large-scale pretraining and zero-shot generalization. In this work, we challenge this convention by showing that discriminative approaches perform substantially better than generative ones on a large number of NLP tasks. Technically, we train a single discriminator to predict whether a text sample comes from the true data distribution, similar to GANs. Since many NLP tasks can be formulated as selecting from a few options, we use this discriminator to predict the concatenation of input and which option has the highest probability of coming from the true data distribution. This simple formulation achieves state-of-the-art zero-shot results on the T0 benchmark, outperforming T0 by 16.0\%, 7.8\%, and 11.5\% respectively on different scales. In the finetuning setting, our approach also achieves new state-of-the-art results on a wide range of NLP tasks, with only 1/4 parameters of previous methods. Meanwhile, our approach requires minimal prompting efforts, which largely improves robustness and is essential for real-world applications. Furthermore, we also jointly train a generalized UD in combination with generative tasks, which maintains its advantage on discriminative tasks and simultaneously works on generative tasks.
comment: ACL 2023 main conference (Long paper)
♻ ☆ PDFVQA: A New Dataset for Real-World VQA on PDF Documents ECML-PKDD 2023
Document-based Visual Question Answering examines the document understanding of document images in conditions of natural language questions. We proposed a new document-based VQA dataset, PDF-VQA, to comprehensively examine the document understanding from various aspects, including document element recognition, document layout structural understanding as well as contextual understanding and key information extraction. Our PDF-VQA dataset extends the current scale of document understanding that limits on the single document page to the new scale that asks questions over the full document of multiple pages. We also propose a new graph-based VQA model that explicitly integrates the spatial and hierarchically structural relationships between different document elements to boost the document structural understanding. The performances are compared with several baselines over different question types and tasks\footnote{The full dataset will be released after paper acceptance.
comment: Accepted by ECML-PKDD 2023
♻ ☆ Controllable Dialogue Simulation with In-Context Learning EMNLP 2022
Building dialogue systems requires a large corpus of annotated dialogues. Such datasets are usually created via crowdsourcing, which is expensive and time-consuming. In this paper, we propose \textsc{Dialogic}, a novel dialogue simulation method based on large language model in-context learning to automate dataset creation. Seeded with a few annotated dialogues, \textsc{Dialogic} automatically selects in-context examples for demonstration and prompts GPT-3 to generate new dialogues and annotations in a controllable way. Our method can rapidly expand a small set of dialogue data with minimum or zero \textit{human involvement} and \textit{parameter update} and is thus much more cost-efficient and time-saving than crowdsourcing. Experimental results on the MultiWOZ dataset demonstrate that training a model on the simulated dialogues leads to even better performance than using the same amount of human-generated dialogues under the challenging low-resource settings, with as few as 85 dialogues as a seed. When enough data is available, our method can still serve as an effective data augmentation method. Human evaluation results also show that our simulated dialogues have near-human fluency and annotation accuracy. The code and data are available at \textbf{\url{https://github.com/Leezekun/dialogic}}.
comment: EMNLP 2022 Findings, code and data are available at https://github.com/Leezekun/dialogic
♻ ☆ Zero-Shot Prompting for Implicit Intent Prediction and Recommendation with Commonsense Reasoning
Intelligent virtual assistants are currently designed to perform tasks or services explicitly mentioned by users, so multiple related domains or tasks need to be performed one by one through a long conversation with many explicit intents. Instead, human assistants are capable of reasoning (multiple) implicit intents based on user utterances via commonsense knowledge, reducing complex interactions and improving practicality. Therefore, this paper proposes a framework of multi-domain dialogue systems, which can automatically infer implicit intents based on user utterances and then perform zero-shot prompting using a large pre-trained language model to trigger suitable single task-oriented bots. The proposed framework is demonstrated effective to realize implicit intents and recommend associated bots in a zero-shot manner.
♻ ☆ A logical word embedding for learning grammar
We introduce the logical grammar emdebbing (LGE), a model inspired by pregroup grammars and categorial grammars to enable unsupervised inference of lexical categories and syntactic rules from a corpus of text. LGE produces comprehensible output summarizing its inferences, has a completely transparent process for producing novel sentences, and can learn from as few as a hundred sentences.
♻ ☆ Retrieval-Augmented Multimodal Language Modeling ICML 2023
Recent multimodal models such as DALL-E and CM3 have achieved remarkable progress in text-to-image and image-to-text generation. However, these models store all learned knowledge (e.g., the appearance of the Eiffel Tower) in the model parameters, requiring increasingly larger models and training data to capture more knowledge. To integrate knowledge in a more scalable and modular way, we propose a retrieval-augmented multimodal model, which enables a base multimodal model (generator) to refer to relevant text and images fetched by a retriever from external memory (e.g., documents on the web). Specifically, for the retriever, we use a pretrained CLIP, and for the generator, we train a CM3 Transformer on the LAION dataset. Our resulting model, named Retrieval-Augmented CM3 (RA-CM3), is the first multimodal model that can retrieve and generate both text and images. We show that RA-CM3 significantly outperforms baseline multimodal models such as DALL-E and CM3 on both image and caption generation tasks (12 FID and 17 CIDEr improvements on MS-COCO), while requiring much less compute for training (<30% of DALL-E). Moreover, we show that RA-CM3 exhibits novel capabilities, such as faithful image generation and multimodal in-context learning (e.g., image generation from demonstrations).
comment: Published at ICML 2023. Blog post available at https://cs.stanford.edu/~myasu/blog/racm3/
♻ ☆ Protecting Language Generation Models via Invisible Watermarking ICML 2023
Language generation models have been an increasingly powerful enabler for many applications. Many such models offer free or affordable API access, which makes them potentially vulnerable to model extraction attacks through distillation. To protect intellectual property (IP) and ensure fair use of these models, various techniques such as lexical watermarking and synonym replacement have been proposed. However, these methods can be nullified by obvious countermeasures such as "synonym randomization". To address this issue, we propose GINSEW, a novel method to protect text generation models from being stolen through distillation. The key idea of our method is to inject secret signals into the probability vector of the decoding steps for each target token. We can then detect the secret message by probing a suspect model to tell if it is distilled from the protected one. Experimental results show that GINSEW can effectively identify instances of IP infringement with minimal impact on the generation quality of protected APIs. Our method demonstrates an absolute improvement of 19 to 29 points on mean average precision (mAP) in detecting suspects compared to previous methods against watermark removal attacks.
comment: ICML 2023
♻ ☆ Fine-Tuning Language Models with Advantage-Induced Policy Alignment
Reinforcement learning from human feedback (RLHF) has emerged as a reliable approach to aligning large language models (LLMs) to human preferences. Among the plethora of RLHF techniques, proximal policy optimization (PPO) is of the most widely used methods. Despite its popularity, however, PPO may suffer from mode collapse, instability, and poor sample efficiency. We show that these issues can be alleviated by a novel algorithm that we refer to as Advantage-Induced Policy Alignment (APA), which leverages a squared error loss function based on the estimated advantages. We demonstrate empirically that APA consistently outperforms PPO in language tasks by a large margin, when a separate reward model is employed as the evaluator. In addition, compared with PPO, APA offers a more stable form of control over the deviation from the model's initial policy, ensuring that the model improves its performance without collapsing to deterministic output. In addition to empirical results, we also provide a theoretical justification supporting the design of our loss function.
♻ ☆ People and Places of Historical Europe: Bootstrapping Annotation Pipeline and a New Corpus of Named Entities in Late Medieval Texts ACL 2023
Although pre-trained named entity recognition (NER) models are highly accurate on modern corpora, they underperform on historical texts due to differences in language OCR errors. In this work, we develop a new NER corpus of 3.6M sentences from late medieval charters written mainly in Czech, Latin, and German. We show that we can start with a list of known historical figures and locations and an unannotated corpus of historical texts, and use information retrieval techniques to automatically bootstrap a NER-annotated corpus. Using our corpus, we train a NER model that achieves entity-level Precision of 72.81-93.98% with 58.14-81.77% Recall on a manually-annotated test dataset. Furthermore, we show that using a weighted loss function helps to combat class imbalance in token classification tasks. To make it easy for others to reproduce and build upon our work, we publicly release our corpus, models, and experimental code.
comment: To appear in the Findings of the Association for Computational Linguistics: ACL 2023
♻ ☆ Privacy-Preserving Domain Adaptation of Semantic Parsers
Task-oriented dialogue systems often assist users with personal or confidential matters. For this reason, the developers of such a system are generally prohibited from observing actual usage. So how can they know where the system is failing and needs more training data or new functionality? In this work, we study ways in which realistic user utterances can be generated synthetically, to help increase the linguistic and functional coverage of the system, without compromising the privacy of actual users. To this end, we propose a two-stage Differentially Private (DP) generation method which first generates latent semantic parses, and then generates utterances based on the parses. Our proposed approach improves MAUVE by 2.5X and parse tree function type overlap by 1.3X relative to current approaches for private synthetic data generation, improving both on fluency and semantic coverage. We further validate our approach on a realistic domain adaptation task of adding new functionality from private user data to a semantic parser, and show overall gains of 8.5% points in accuracy with the new feature.
♻ ☆ FEED PETs: Further Experimentation and Expansion on the Disambiguation of Potentially Euphemistic Terms
Transformers have been shown to work well for the task of English euphemism disambiguation, in which a potentially euphemistic term (PET) is classified as euphemistic or non-euphemistic in a particular context. In this study, we expand on the task in two ways. First, we annotate PETs for vagueness, a linguistic property associated with euphemisms, and find that transformers are generally better at classifying vague PETs, suggesting linguistic differences in the data that impact performance. Second, we present novel euphemism corpora in three different languages: Yoruba, Spanish, and Mandarin Chinese. We perform euphemism disambiguation experiments in each language using multilingual transformer models mBERT and XLM-RoBERTa, establishing preliminary results from which to launch future work.
Computer Vision and Pattern Recognition 141
☆ SAM3D: Segment Anything in 3D Scenes
In this work, we propose SAM3D, a novel framework that is able to predict masks in 3D point clouds by leveraging the Segment-Anything Model (SAM) in RGB images without further training or finetuning. For a point cloud of a 3D scene with posed RGB images, we first predict segmentation masks of RGB images with SAM, and then project the 2D masks into the 3D points. Later, we merge the 3D masks iteratively with a bottom-up merging approach. At each step, we merge the point cloud masks of two adjacent frames with the bidirectional merging approach. In this way, the 3D masks predicted from different frames are gradually merged into the 3D masks of the whole 3D scene. Finally, we can optionally ensemble the result from our SAM3D with the over-segmentation results based on the geometric information of the 3D scenes. Our approach is experimented with ScanNet dataset and qualitative results demonstrate that our SAM3D achieves reasonable and fine-grained 3D segmentation results without any training or finetuning of SAM.
comment: Technical Report. The code is released at https://github.com/Pointcept/SegmentAnything3D
☆ Towards Label-free Scene Understanding by Vision Foundation Models
Vision foundation models such as Contrastive Vision-Language Pre-training (CLIP) and Segment Anything (SAM) have demonstrated impressive zero-shot performance on image classification and segmentation tasks. However, the incorporation of CLIP and SAM for label-free scene understanding has yet to be explored. In this paper, we investigate the potential of vision foundation models in enabling networks to comprehend 2D and 3D worlds without labelled data. The primary challenge lies in effectively supervising networks under extremely noisy pseudo labels, which are generated by CLIP and further exacerbated during the propagation from the 2D to the 3D domain. To tackle these challenges, we propose a novel Cross-modality Noisy Supervision (CNS) method that leverages the strengths of CLIP and SAM to supervise 2D and 3D networks simultaneously. In particular, we introduce a prediction consistency regularization to co-train 2D and 3D networks, then further impose the networks' latent space consistency using the SAM's robust feature representation. Experiments conducted on diverse indoor and outdoor datasets demonstrate the superior performance of our method in understanding 2D and 3D open environments. Our 2D and 3D network achieves label-free semantic segmentation with 28.4% and 33.5% mIoU on ScanNet, improving 4.7% and 7.9%, respectively. And for nuScenes dataset, our performance is 26.8% with an improvement of 6%. Code will be released (https://github.com/runnanchen/Label-Free-Scene-Understanding).
☆ Emergent Correspondence from Image Diffusion
Finding correspondences between images is a fundamental problem in computer vision. In this paper, we show that correspondence emerges in image diffusion models without any explicit supervision. We propose a simple strategy to extract this implicit knowledge out of diffusion networks as image features, namely DIffusion FeaTures (DIFT), and use them to establish correspondences between real images. Without any additional fine-tuning or supervision on the task-specific data or annotations, DIFT is able to outperform both weakly-supervised methods and competitive off-the-shelf features in identifying semantic, geometric, and temporal correspondences. Particularly for semantic correspondence, DIFT from Stable Diffusion is able to outperform DINO and OpenCLIP by 19 and 14 accuracy points respectively on the challenging SPair-71k benchmark. It even outperforms the state-of-the-art supervised methods on 9 out of 18 categories while remaining on par for the overall performance. Project page: https://diffusionfeatures.github.io
comment: Project page: https://diffusionfeatures.github.io
☆ Conditional Diffusion Models for Weakly Supervised Medical Image Segmentation
Recent advances in denoising diffusion probabilistic models have shown great success in image synthesis tasks. While there are already works exploring the potential of this powerful tool in image semantic segmentation, its application in weakly supervised semantic segmentation (WSSS) remains relatively under-explored. Observing that conditional diffusion models (CDM) is capable of generating images subject to specific distributions, in this work, we utilize category-aware semantic information underlied in CDM to get the prediction mask of the target object with only image-level annotations. More specifically, we locate the desired class by approximating the derivative of the output of CDM w.r.t the input condition. Our method is different from previous diffusion model methods with guidance from an external classifier, which accumulates noises in the background during the reconstruction process. Our method outperforms state-of-the-art CAM and diffusion model methods on two public medical image segmentation datasets, which demonstrates that CDM is a promising tool in WSSS. Also, experiment shows our method is more time-efficient than existing diffusion model methods, making it practical for wider applications.
☆ Learning with a Mole: Transferable latent spatial representations for navigation without reconstruction
Agents navigating in 3D environments require some form of memory, which should hold a compact and actionable representation of the history of observations useful for decision taking and planning. In most end-to-end learning approaches the representation is latent and usually does not have a clearly defined interpretation, whereas classical robotics addresses this with scene reconstruction resulting in some form of map, usually estimated with geometry and sensor models and/or learning. In this work we propose to learn an actionable representation of the scene independently of the targeted downstream task and without explicitly optimizing reconstruction. The learned representation is optimized by a blind auxiliary agent trained to navigate with it on multiple short sub episodes branching out from a waypoint and, most importantly, without any direct visual observation. We argue and show that the blindness property is important and forces the (trained) latent representation to be the only means for planning. With probing experiments we show that the learned representation optimizes navigability and not reconstruction. On downstream tasks we show that it is robust to changes in distribution, in particular the sim2real gap, which we evaluate with a real physical robot in a real office building, significantly improving performance.
☆ Learning Human Mesh Recovery in 3D Scenes CVPR 2023
We present a novel method for recovering the absolute pose and shape of a human in a pre-scanned scene given a single image. Unlike previous methods that perform sceneaware mesh optimization, we propose to first estimate absolute position and dense scene contacts with a sparse 3D CNN, and later enhance a pretrained human mesh recovery network by cross-attention with the derived 3D scene cues. Joint learning on images and scene geometry enables our method to reduce the ambiguity caused by depth and occlusion, resulting in more reasonable global postures and contacts. Encoding scene-aware cues in the network also allows the proposed method to be optimization-free, and opens up the opportunity for real-time applications. The experiments show that the proposed network is capable of recovering accurate and physically-plausible meshes by a single forward pass and outperforms state-of-the-art methods in terms of both accuracy and speed.
comment: Accepted to CVPR 2023. Project page: https://zju3dv.github.io/sahmr/
☆ Atrial Septal Defect Detection in Children Based on Ultrasound Video Using Multiple Instances Learning
Purpose: Congenital heart defect (CHD) is the most common birth defect. Thoracic echocardiography (TTE) can provide sufficient cardiac structure information, evaluate hemodynamics and cardiac function, and is an effective method for atrial septal defect (ASD) examination. This paper aims to study a deep learning method based on cardiac ultrasound video to assist in ASD diagnosis. Materials and methods: We select two standard views of the atrial septum (subAS) and low parasternal four-compartment view (LPS4C) as the two views to identify ASD. We enlist data from 300 children patients as part of a double-blind experiment for five-fold cross-validation to verify the performance of our model. In addition, data from 30 children patients (15 positives and 15 negatives) are collected for clinician testing and compared to our model test results (these 30 samples do not participate in model training). We propose an echocardiography video-based atrial septal defect diagnosis system. In our model, we present a block random selection, maximal agreement decision and frame sampling strategy for training and testing respectively, resNet18 and r3D networks are used to extract the frame features and aggregate them to build a rich video-level representation. Results: We validate our model using our private dataset by five-cross validation. For ASD detection, we achieve 89.33 AUC, 84.95 accuracy, 85.70 sensitivity, 81.51 specificity and 81.99 F1 score. Conclusion: The proposed model is multiple instances learning-based deep learning model for video atrial septal defect detection which effectively improves ASD detection accuracy when compared to the performances of previous networks and clinical doctors.
GEO-Bench: Toward Foundation Models for Earth Monitoring
Recent progress in self-supervision has shown that pre-training large neural networks on vast amounts of unsupervised data can lead to substantial increases in generalization to downstream tasks. Such models, recently coined foundation models, have been transformational to the field of natural language processing. Variants have also been proposed for image data, but their applicability to remote sensing tasks is limited. To stimulate the development of foundation models for Earth monitoring, we propose a benchmark comprised of six classification and six segmentation tasks, which were carefully curated and adapted to be both relevant to the field and well-suited for model evaluation. We accompany this benchmark with a robust methodology for evaluating models and reporting aggregated results to enable a reliable assessment of progress. Finally, we report results for 20 baselines to gain information about the performance of existing models. We believe that this benchmark will be a driver of progress across a variety of Earth monitoring tasks.
☆ X-Align++: cross-modal cross-view alignment for Bird's-eye-view segmentation
Bird's-eye-view (BEV) grid is a typical representation of the perception of road components, e.g., drivable area, in autonomous driving. Most existing approaches rely on cameras only to perform segmentation in BEV space, which is fundamentally constrained by the absence of reliable depth information. The latest works leverage both camera and LiDAR modalities but suboptimally fuse their features using simple, concatenation-based mechanisms. In this paper, we address these problems by enhancing the alignment of the unimodal features in order to aid feature fusion, as well as enhancing the alignment between the cameras' perspective view (PV) and BEV representations. We propose X-Align, a novel end-to-end cross-modal and cross-view learning framework for BEV segmentation consisting of the following components: (i) a novel Cross-Modal Feature Alignment (X-FA) loss, (ii) an attention-based Cross-Modal Feature Fusion (X-FF) module to align multi-modal BEV features implicitly, and (iii) an auxiliary PV segmentation branch with Cross-View Segmentation Alignment (X-SA) losses to improve the PV-to-BEV transformation. We evaluate our proposed method across two commonly used benchmark datasets, i.e., nuScenes and KITTI-360. Notably, X-Align significantly outperforms the state-of-the-art by 3 absolute mIoU points on nuScenes. We also provide extensive ablation studies to demonstrate the effectiveness of the individual components.
comment: Accepted for publication at Springer Machine Vision and Applications Journal. The Version of Record of this article is published in Machine Vision and Applications Journal, and is available online at https://doi.org/10.1007/s00138-023-01400-7. arXiv admin note: substantial text overlap with arXiv:2210.06778
☆ Learning to Ground Instructional Articles in Videos through Narrations
In this paper we present an approach for localizing steps of procedural activities in narrated how-to videos. To deal with the scarcity of labeled data at scale, we source the step descriptions from a language knowledge base (wikiHow) containing instructional articles for a large variety of procedural tasks. Without any form of manual supervision, our model learns to temporally ground the steps of procedural articles in how-to videos by matching three modalities: frames, narrations, and step descriptions. Specifically, our method aligns steps to video by fusing information from two distinct pathways: i) {\em direct} alignment of step descriptions to frames, ii) {\em indirect} alignment obtained by composing steps-to-narrations with narrations-to-video correspondences. Notably, our approach performs global temporal grounding of all steps in an article at once by exploiting order information, and is trained with step pseudo-labels which are iteratively refined and aggressively filtered. In order to validate our model we introduce a new evaluation benchmark -- HT-Step -- obtained by manually annotating a 124-hour subset of HowTo100M\footnote{A test server is accessible at \url{https://eval.ai/web/challenges/challenge-page/2082}.} with steps sourced from wikiHow articles. Experiments on this benchmark as well as zero-shot evaluations on CrossTask demonstrate that our multi-modality alignment yields dramatic gains over several baselines and prior works. Finally, we show that our inner module for matching narration-to-video outperforms by a large margin the state of the art on the HTM-Align narration-video alignment benchmark.
comment: 17 pages, 4 figures and 10 tables
☆ Performance-optimized deep neural networks are evolving into worse models of inferotemporal visual cortex
One of the most impactful findings in computational neuroscience over the past decade is that the object recognition accuracy of deep neural networks (DNNs) correlates with their ability to predict neural responses to natural images in the inferotemporal (IT) cortex. This discovery supported the long-held theory that object recognition is a core objective of the visual cortex, and suggested that more accurate DNNs would serve as better models of IT neuron responses to images. Since then, deep learning has undergone a revolution of scale: billion parameter-scale DNNs trained on billions of images are rivaling or outperforming humans at visual tasks including object recognition. Have today's DNNs become more accurate at predicting IT neuron responses to images as they have grown more accurate at object recognition? Surprisingly, across three independent experiments, we find this is not the case. DNNs have become progressively worse models of IT as their accuracy has increased on ImageNet. To understand why DNNs experience this trade-off and evaluate if they are still an appropriate paradigm for modeling the visual system, we turn to recordings of IT that capture spatially resolved maps of neuronal activity elicited by natural images. These neuronal activity maps reveal that DNNs trained on ImageNet learn to rely on different visual features than those encoded by IT and that this problem worsens as their accuracy increases. We successfully resolved this issue with the neural harmonizer, a plug-and-play training routine for DNNs that aligns their learned representations with humans. Our results suggest that harmonized DNNs break the trade-off between ImageNet accuracy and neural prediction accuracy that assails current DNNs and offer a path to more accurate models of biological vision.
☆ Newly Formed Cities: an AI Curation
Art curatorial processes are characterized by the presentation of a collection of artworks in a knowledgeable way. Machine processes are characterized by their capacity to manage and analyze large amounts of data. This paper envisages machine curation and audience interaction as a means to explore the implications of contemporary AI models for the curatorial world. This project was developed for the occasion of the 2023 Helsinki Art Biennial, entitled New Directions May Emerge. We use the Helsinki Art Museum (HAM) collection to re-imagine the city of Helsinki through the lens of machine perception. We use visual-textual models to place artworks currently hosted inside the museum in outdoor public spaces of the city, assigning fictional coordinates based on similarity scores. Synthetic 360{\deg} art panoramas are generated using diffusion-based models to propose a machinic visual style guided by the artworks. The result of this project will be virtually presented as a web-based installation, where such a re-contextualization allows the navigation of an alternative version of the city while exploring its artistic heritage. Finally, we discuss our contributions to machine curation and the ethical implications that such a process entails. The web-based installation is available at this link: http://newlyformedcity.com/.
☆ Towards Scalable Multi-View Reconstruction of Geometry and Materials
In this paper, we propose a novel method for joint recovery of camera pose, object geometry and spatially-varying Bidirectional Reflectance Distribution Function (svBRDF) of 3D scenes that exceed object-scale and hence cannot be captured with stationary light stages. The input are high-resolution RGB-D images captured by a mobile, hand-held capture system with point lights for active illumination. Compared to previous works that jointly estimate geometry and materials from a hand-held scanner, we formulate this problem using a single objective function that can be minimized using off-the-shelf gradient-based solvers. To facilitate scalability to large numbers of observation views and optimization variables, we introduce a distributed optimization algorithm that reconstructs 2.5D keyframe-based representations of the scene. A novel multi-view consistency regularizer effectively synchronizes neighboring keyframes such that the local optimization results allow for seamless integration into a globally consistent 3D model. We provide a study on the importance of each component in our formulation and show that our method compares favorably to baselines. We further demonstrate that our method accurately reconstructs various objects and materials and allows for expansion to spatially larger scenes. We believe that this work represents a significant step towards making geometry and material estimation from hand-held scanners scalable.
☆ Modality-Agnostic Learning for Medical Image Segmentation Using Multi-modality Self-distillation
Medical image segmentation of tumors and organs at risk is a time-consuming yet critical process in the clinic that utilizes multi-modality imaging (e.g, different acquisitions, data types, and sequences) to increase segmentation precision. In this paper, we propose a novel framework, Modality-Agnostic learning through Multi-modality Self-dist-illation (MAG-MS), to investigate the impact of input modalities on medical image segmentation. MAG-MS distills knowledge from the fusion of multiple modalities and applies it to enhance representation learning for individual modalities. Thus, it provides a versatile and efficient approach to handle limited modalities during testing. Our extensive experiments on benchmark datasets demonstrate the high efficiency of MAG-MS and its superior segmentation performance than current state-of-the-art methods. Furthermore, using MAG-MS, we provide valuable insight and guidance on selecting input modalities for medical image segmentation tasks.
☆ Towards Visual Foundational Models of Physical Scenes
We describe a first step towards learning general-purpose visual representations of physical scenes using only image prediction as a training criterion. To do so, we first define "physical scene" and show that, even though different agents may maintain different representations of the same scene, the underlying physical scene that can be inferred is unique. Then, we show that NeRFs cannot represent the physical scene, as they lack extrapolation mechanisms. Those, however, could be provided by Diffusion Models, at least in theory. To test this hypothesis empirically, NeRFs can be combined with Diffusion Models, a process we refer to as NeRF Diffusion, used as unsupervised representations of the physical scene. Our analysis is limited to visual data, without external grounding mechanisms that can be provided by independent sensory modalities.
comment: TLDR: Physical scenes are equivalence classes of sufficient statistics, and can be inferred uniquely by any agent measuring the same finite data; We formalize and implement an approach to representation learning that overturns "naive realism" in favor of an analytical approach of Russell and Koenderink. NeRFs cannot capture the physical scenes, but combined with Diffusion Models they can
☆ Deep Learning-Enabled Sleep Staging From Vital Signs and Activity Measured Using a Near-Infrared Video Camera CVPR 2023
Conventional sleep monitoring is time-consuming, expensive and uncomfortable, requiring a large number of contact sensors to be attached to the patient. Video data is commonly recorded as part of a sleep laboratory assessment. If accurate sleep staging could be achieved solely from video, this would overcome many of the problems of traditional methods. In this work we use heart rate, breathing rate and activity measures, all derived from a near-infrared video camera, to perform sleep stage classification. We use a deep transfer learning approach to overcome data scarcity, by using an existing contact-sensor dataset to learn effective representations from the heart and breathing rate time series. Using a dataset of 50 healthy volunteers, we achieve an accuracy of 73.4\% and a Cohen's kappa of 0.61 in four-class sleep stage classification, establishing a new state-of-the-art for video-based sleep staging.
comment: Accepted to the 6th International Workshop on Computer Vision for Physiological Measurement (CVPM) at CVPR 2023. 10 pages, 12 figures, 5 tables
☆ YONA: You Only Need One Adjacent Reference-frame for Accurate and Fast Video Polyp Detection MICCAI2023
Accurate polyp detection is essential for assisting clinical rectal cancer diagnoses. Colonoscopy videos contain richer information than still images, making them a valuable resource for deep learning methods. Great efforts have been made to conduct video polyp detection through multi-frame temporal/spatial aggregation. However, unlike common fixed-camera video, the camera-moving scene in colonoscopy videos can cause rapid video jitters, leading to unstable training for existing video detection models. Additionally, the concealed nature of some polyps and the complex background environment further hinder the performance of existing video detectors. In this paper, we propose the \textbf{YONA} (\textbf{Y}ou \textbf{O}nly \textbf{N}eed one \textbf{A}djacent Reference-frame) method, an efficient end-to-end training framework for video polyp detection. YONA fully exploits the information of one previous adjacent frame and conducts polyp detection on the current frame without multi-frame collaborations. Specifically, for the foreground, YONA adaptively aligns the current frame's channel activation patterns with its adjacent reference frames according to their foreground similarity. For the background, YONA conducts background dynamic alignment guided by inter-frame difference to eliminate the invalid features produced by drastic spatial jitters. Moreover, YONA applies cross-frame contrastive learning during training, leveraging the ground truth bounding box to improve the model's perception of polyp and background. Quantitative and qualitative experiments on three public challenging benchmarks demonstrate that our proposed YONA outperforms previous state-of-the-art competitors by a large margin in both accuracy and speed.
comment: 11 pages, 3 figures, Accepted by MICCAI2023
☆ Human-imperceptible, Machine-recognizable Images
Massive human-related data is collected to train neural networks for computer vision tasks. A major conflict is exposed relating to software engineers between better developing AI systems and distancing from the sensitive training data. To reconcile this conflict, this paper proposes an efficient privacy-preserving learning paradigm, where images are first encrypted to become ``human-imperceptible, machine-recognizable'' via one of the two encryption strategies: (1) random shuffling to a set of equally-sized patches and (2) mixing-up sub-patches of the images. Then, minimal adaptations are made to vision transformer to enable it to learn on the encrypted images for vision tasks, including image classification and object detection. Extensive experiments on ImageNet and COCO show that the proposed paradigm achieves comparable accuracy with the competitive methods. Decrypting the encrypted images requires solving an NP-hard jigsaw puzzle or an ill-posed inverse problem, which is empirically shown intractable to be recovered by various attackers, including the powerful vision transformer-based attacker. We thus show that the proposed paradigm can ensure the encrypted images have become human-imperceptible while preserving machine-recognizable information. The code is available at \url{https://github.com/FushengHao/PrivacyPreservingML.}
☆ PQM: A Point Quality Evaluation Metric for Dense Maps
LiDAR-based mapping/reconstruction are important for various applications, but evaluating the quality of the dense maps they produce is challenging. The current methods have limitations, including the inability to capture completeness, structural information, and local variations in error. In this paper, we propose a novel point quality evaluation metric (PQM) that consists of four sub-metrics to provide a more comprehensive evaluation of point cloud quality. The completeness sub-metric evaluates the proportion of missing data, the artifact score sub-metric recognizes and characterizes artifacts, the accuracy sub-metric measures registration accuracy, and the resolution sub-metric quantifies point cloud density. Through an ablation study using a prototype dataset, we demonstrate the effectiveness of each of the sub-metrics and compare them to popular point cloud distance measures. Using three LiDAR SLAM systems to generate maps, we evaluate their output map quality and demonstrate the metrics robustness to noise and artifacts. Our implementation of PQM, datasets and detailed documentation on how to integrate with your custom dense mapping pipeline can be found at github.com/droneslab/pqm
☆ Supervised Knowledge May Hurt Novel Class Discovery Performance
Novel class discovery (NCD) aims to infer novel categories in an unlabeled dataset by leveraging prior knowledge of a labeled set comprising disjoint but related classes. Given that most existing literature focuses primarily on utilizing supervised knowledge from a labeled set at the methodology level, this paper considers the question: Is supervised knowledge always helpful at different levels of semantic relevance? To proceed, we first establish a novel metric, so-called transfer flow, to measure the semantic similarity between labeled/unlabeled datasets. To show the validity of the proposed metric, we build up a large-scale benchmark with various degrees of semantic similarities between labeled/unlabeled datasets on ImageNet by leveraging its hierarchical class structure. The results based on the proposed benchmark show that the proposed transfer flow is in line with the hierarchical class structure; and that NCD performance is consistent with the semantic similarities (measured by the proposed metric). Next, by using the proposed transfer flow, we conduct various empirical experiments with different levels of semantic similarity, yielding that supervised knowledge may hurt NCD performance. Specifically, using supervised information from a low-similarity labeled set may lead to a suboptimal result as compared to using pure self-supervised knowledge. These results reveal the inadequacy of the existing NCD literature which usually assumes that supervised knowledge is beneficial. Finally, we develop a pseudo-version of the transfer flow as a practical reference to decide if supervised knowledge should be used in NCD. Its effectiveness is supported by our empirical studies, which show that the pseudo transfer flow (with or without supervised knowledge) is consistent with the corresponding accuracy based on various datasets. Code is released at https://github.com/J-L-O/SK-Hurt-NCD
comment: TMLR 2023 accepted paper. arXiv admin note: substantial text overlap with arXiv:2209.09120
☆ Single-Shot Global Localization via Graph-Theoretic Correspondence Matching
This paper describes a method of global localization based on graph-theoretic association of instances between a query and the prior map. The proposed framework employs correspondence matching based on the maximum clique problem (MCP). The framework is potentially applicable to other map and/or query modalities thanks to the graph-based abstraction of the problem, while many of existing global localization methods rely on a query and the dataset in the same modality. We implement it with a semantically labeled 3D point cloud map, and a semantic segmentation image as a query. Leveraging the graph-theoretic framework, the proposed method realizes global localization exploiting only the map and the query. The method shows promising results on multiple large-scale simulated maps of urban scenes.
☆ Mutual Information Regularization for Weakly-supervised RGB-D Salient Object Detection
In this paper, we present a weakly-supervised RGB-D salient object detection model via scribble supervision. Specifically, as a multimodal learning task, we focus on effective multimodal representation learning via inter-modal mutual information regularization. In particular, following the principle of disentangled representation learning, we introduce a mutual information upper bound with a mutual information minimization regularizer to encourage the disentangled representation of each modality for salient object detection. Based on our multimodal representation learning framework, we introduce an asymmetric feature extractor for our multimodal data, which is proven more effective than the conventional symmetric backbone setting. We also introduce multimodal variational auto-encoder as stochastic prediction refinement techniques, which takes pseudo labels from the first training stage as supervision and generates refined prediction. Experimental results on benchmark RGB-D salient object detection datasets verify both effectiveness of our explicit multimodal disentangled representation learning method and the stochastic prediction refinement strategy, achieving comparable performance with the state-of-the-art fully supervised models. Our code and data are available at: https://github.com/baneitixiaomai/MIRV.
comment: IEEE Transactions on Circuits and Systems for Video Technology 2023
☆ Digitization of Pathology Labs: A Review of Lessons Learned
Pathology laboratories are increasingly using digital workflows. This has the potential of increasing lab efficiency, but the digitization process also involves major challenges. Several reports have been published describing the individual experiences of specific laboratories with the digitization process. However, a comprehensive overview of the lessons learned is still lacking. We provide an overview of the lessons learned for different aspects of the digitization process, including digital case management, digital slide reading, and computer-aided slide reading. We also cover metrics used for monitoring performance and pitfalls and corresponding values observed in practice. The overview is intended to help pathologists, IT decision-makers, and administrators to benefit from the experiences of others and to implement the digitization process in an optimal way to make their own laboratory future-proof.
comment: 22 pages, 1 figure
☆ Human-Object Interaction Prediction in Videos through Gaze Following
Understanding the human-object interactions (HOIs) from a video is essential to fully comprehend a visual scene. This line of research has been addressed by detecting HOIs from images and lately from videos. However, the video-based HOI anticipation task in the third-person view remains understudied. In this paper, we design a framework to detect current HOIs and anticipate future HOIs in videos. We propose to leverage human gaze information since people often fixate on an object before interacting with it. These gaze features together with the scene contexts and the visual appearances of human-object pairs are fused through a spatio-temporal transformer. To evaluate the model in the HOI anticipation task in a multi-person scenario, we propose a set of person-wise multi-label metrics. Our model is trained and validated on the VidHOI dataset, which contains videos capturing daily life and is currently the largest video HOI dataset. Experimental results in the HOI detection task show that our approach improves the baseline by a great margin of 36.3% relatively. Moreover, we conduct an extensive ablation study to demonstrate the effectiveness of our modifications and extensions to the spatio-temporal transformer. Our code is publicly available on https://github.com/nizhf/hoi-prediction-gaze-transformer.
comment: Accepted by CVIU https://doi.org/10.1016/j.cviu.2023.103741
☆ Emotional Talking Head Generation based on Memory-Sharing and Attention-Augmented Networks
Given an audio clip and a reference face image, the goal of the talking head generation is to generate a high-fidelity talking head video. Although some audio-driven methods of generating talking head videos have made some achievements in the past, most of them only focused on lip and audio synchronization and lack the ability to reproduce the facial expressions of the target person. To this end, we propose a talking head generation model consisting of a Memory-Sharing Emotion Feature extractor (MSEF) and an Attention-Augmented Translator based on U-net (AATU). Firstly, MSEF can extract implicit emotional auxiliary features from audio to estimate more accurate emotional face landmarks.~Secondly, AATU acts as a translator between the estimated landmarks and the photo-realistic video frames. Extensive qualitative and quantitative experiments have shown the superiority of the proposed method to the previous works. Codes will be made publicly available.
☆ RDFC-GAN: RGB-Depth Fusion CycleGAN for Indoor Depth Completion CVPR 2022
The raw depth image captured by indoor depth sensors usually has an extensive range of missing depth values due to inherent limitations such as the inability to perceive transparent objects and the limited distance range. The incomplete depth map with missing values burdens many downstream vision tasks, and a rising number of depth completion methods have been proposed to alleviate this issue. While most existing methods can generate accurate dense depth maps from sparse and uniformly sampled depth maps, they are not suitable for complementing large contiguous regions of missing depth values, which is common and critical in images captured in indoor environments. To overcome these challenges, we design a novel two-branch end-to-end fusion network named RDFC-GAN, which takes a pair of RGB and incomplete depth images as input to predict a dense and completed depth map. The first branch employs an encoder-decoder structure, by adhering to the Manhattan world assumption and utilizing normal maps from RGB-D information as guidance, to regress the local dense depth values from the raw depth map. In the other branch, we propose an RGB-depth fusion CycleGAN to transfer the RGB image to the fine-grained textured depth map. We adopt adaptive fusion modules named W-AdaIN to propagate the features across the two branches, and we append a confidence fusion head to fuse the two outputs of the branches for the final depth map. Extensive experiments on NYU-Depth V2 and SUN RGB-D demonstrate that our proposed method clearly improves the depth completion performance, especially in a more realistic setting of indoor environments, with the help of our proposed pseudo depth maps in training.
comment: Haowen Wang and Zhengping Che are with equal contributions. Under review. An earlier version has been accepted by CVPR 2022 (arXiv:2203.10856)
☆ An Open Patch Generator based Fingerprint Presentation Attack Detection using Generative Adversarial Network
The low-cost, user-friendly, and convenient nature of Automatic Fingerprint Recognition Systems (AFRS) makes them suitable for a wide range of applications. This spreading use of AFRS also makes them vulnerable to various security threats. Presentation Attack (PA) or spoofing is one of the threats which is caused by presenting a spoof of a genuine fingerprint to the sensor of AFRS. Fingerprint Presentation Attack Detection (FPAD) is a countermeasure intended to protect AFRS against fake or spoof fingerprints created using various fabrication materials. In this paper, we have proposed a Convolutional Neural Network (CNN) based technique that uses a Generative Adversarial Network (GAN) to augment the dataset with spoof samples generated from the proposed Open Patch Generator (OPG). This OPG is capable of generating realistic fingerprint samples which have no resemblance to the existing spoof fingerprint samples generated with other materials. The augmented dataset is fed to the DenseNet classifier which helps in increasing the performance of the Presentation Attack Detection (PAD) module for the various real-world attacks possible with unknown spoof materials. Experimental evaluations of the proposed approach are carried out on the Liveness Detection (LivDet) 2015, 2017, and 2019 competition databases. An overall accuracy of 96.20\%, 94.97\%, and 92.90\% has been achieved on the LivDet 2015, 2017, and 2019 databases, respectively under the LivDet protocol scenarios. The performance of the proposed PAD model is also validated in the cross-material and cross-sensor attack paradigm which further exhibits its capability to be used under real-world attack scenarios.
☆ Human 3D Avatar Modeling with Implicit Neural Representation: A Brief Survey
A human 3D avatar is one of the important elements in the metaverse, and the modeling effect directly affects people's visual experience. However, the human body has a complex topology and diverse details, so it is often expensive, time-consuming, and laborious to build a satisfactory model. Recent studies have proposed a novel method, implicit neural representation, which is a continuous representation method and can describe objects with arbitrary topology at arbitrary resolution. Researchers have applied implicit neural representation to human 3D avatar modeling and obtained more excellent results than traditional methods. This paper comprehensively reviews the application of implicit neural representation in human body modeling. First, we introduce three implicit representations of occupancy field, SDF, and NeRF, and make a classification of the literature investigated in this paper. Then the application of implicit modeling methods in the body, hand, and head are compared and analyzed respectively. Finally, we point out the shortcomings of current work and provide available suggestions for researchers.
comment: A Brief Survey
☆ Scalable Concept Extraction in Industry 4.0
The industry 4.0 is leveraging digital technologies and machine learning techniques to connect and optimize manufacturing processes. Central to this idea is the ability to transform raw data into human understandable knowledge for reliable data-driven decision-making. Convolutional Neural Networks (CNNs) have been instrumental in processing image data, yet, their ``black box'' nature complicates the understanding of their prediction process. In this context, recent advances in the field of eXplainable Artificial Intelligence (XAI) have proposed the extraction and localization of concepts, or which visual cues intervene on the prediction process of CNNs. This paper tackles the application of concept extraction (CE) methods to industry 4.0 scenarios. To this end, we modify a recently developed technique, ``Extracting Concepts with Local Aggregated Descriptors'' (ECLAD), improving its scalability. Specifically, we propose a novel procedure for calculating concept importance, utilizing a wrapper function designed for CNNs. This process is aimed at decreasing the number of times each image needs to be evaluated. Subsequently, we demonstrate the potential of CE methods, by applying them in three industrial use cases. We selected three representative use cases in the context of quality control for material design (tailored textiles), manufacturing (carbon fiber reinforcement), and maintenance (photovoltaic module inspection). In these examples, CE was able to successfully extract and locate concepts directly related to each task. This is, the visual cues related to each concept, coincided with what human experts would use to perform the task themselves, even when the visual cues were entangled between multiple classes. Through empirical results, we show that CE can be applied for understanding CNNs in an industrial context, giving useful insights that can relate to domain knowledge.
☆ SDR-GAIN: A High Real-Time Occluded Pedestrian Pose Completion Method for Autonomous Driving
To mitigate the challenges arising from partial occlusion in human pose keypoint based pedestrian detection methods , we present a novel pedestrian pose keypoint completion method called the separation and dimensionality reduction-based generative adversarial imputation networks (SDR-GAIN) . Firstly, we utilize OpenPose to estimate pedestrian poses in images. Then, we isolate the head and torso keypoints of pedestrians with incomplete keypoints due to occlusion or other factors and perform dimensionality reduction to enhance features and further unify feature distribution. Finally, we introduce two generative models based on the generative adversarial networks (GAN) framework, which incorporate Huber loss, residual structure, and L1 regularization to generate missing parts of the incomplete head and torso pose keypoints of partially occluded pedestrians, resulting in pose completion. Our experiments on MS COCO and JAAD datasets demonstrate that SDR-GAIN outperforms basic GAIN framework, interpolation methods PCHIP and MAkima, machine learning methods k-NN and MissForest in terms of pose completion task. In addition, the runtime of SDR-GAIN is approximately 0.4ms, displaying high real-time performance and significant application value in the field of autonomous driving.
☆ Real-Time Onboard Object Detection for Augmented Reality: Enhancing Head-Mounted Display with YOLOv8
This paper introduces a software architecture for real-time object detection using machine learning (ML) in an augmented reality (AR) environment. Our approach uses the recent state-of-the-art YOLOv8 network that runs onboard on the Microsoft HoloLens 2 head-mounted display (HMD). The primary motivation behind this research is to enable the application of advanced ML models for enhanced perception and situational awareness with a wearable, hands-free AR platform. We show the image processing pipeline for the YOLOv8 model and the techniques used to make it real-time on the resource-limited edge computing platform of the headset. The experimental results demonstrate that our solution achieves real-time processing without needing offloading tasks to the cloud or any other external servers while retaining satisfactory accuracy regarding the usual mAP metric and measured qualitative performance
☆ Expanding Explainability Horizons: A Unified Concept-Based System for Local, Global, and Misclassification Explanations
Explainability of intelligent models has been garnering increasing attention in recent years. Of the various explainability approaches, concept-based techniques are notable for utilizing a set of human-meaningful concepts instead of focusing on individual pixels. However, there is a scarcity of methods that consistently provide both local and global explanations. Moreover, most of the methods have no offer to explain misclassification cases. To address these challenges, our study follows a straightforward yet effective approach. We propose a unified concept-based system, which inputs a number of super-pixelated images into the networks, allowing them to learn better representations of the target's objects as well as the target's concepts. This method automatically learns, scores, and extracts local and global concepts. Our experiments revealed that, in addition to enhancing performance, the models could provide deeper insights into predictions and elucidate false classifications.
☆ A Functional Data Perspective and Baseline On Multi-Layer Out-of-Distribution Detection
A key feature of out-of-distribution (OOD) detection is to exploit a trained neural network by extracting statistical patterns and relationships through the multi-layer classifier to detect shifts in the expected input data distribution. Despite achieving solid results, several state-of-the-art methods rely on the penultimate or last layer outputs only, leaving behind valuable information for OOD detection. Methods that explore the multiple layers either require a special architecture or a supervised objective to do so. This work adopts an original approach based on a functional view of the network that exploits the sample's trajectories through the various layers and their statistical dependencies. It goes beyond multivariate features aggregation and introduces a baseline rooted in functional anomaly detection. In this new framework, OOD detection translates into detecting samples whose trajectories differ from the typical behavior characterized by the training set. We validate our method and empirically demonstrate its effectiveness in OOD detection compared to strong state-of-the-art baselines on computer vision benchmarks.
☆ Recognize Anything: A Strong Image Tagging Model
We present the Recognize Anything Model (RAM): a strong foundation model for image tagging. RAM can recognize any common category with high accuracy. RAM introduces a new paradigm for image tagging, leveraging large-scale image-text pairs for training instead of manual annotations. The development of RAM comprises four key steps. Firstly, annotation-free image tags are obtained at scale through automatic text semantic parsing. Subsequently, a preliminary model is trained for automatic annotation by unifying the caption and tagging tasks, supervised by the original texts and parsed tags, respectively. Thirdly, a data engine is employed to generate additional annotations and clean incorrect ones. Lastly, the model is retrained with the processed data and fine-tuned using a smaller but higher-quality dataset. We evaluate the tagging capabilities of RAM on numerous benchmarks and observe impressive zero-shot performance, significantly outperforming CLIP and BLIP. Remarkably, RAM even surpasses the fully supervised manners and exhibits competitive performance with the Google API. We are releasing the RAM at \url{https://recognize-anything.github.io/} to foster the advancements of large models in computer vision.
☆ Curriculum-Based Augmented Fourier Domain Adaptation for Robust Medical Image Segmentation
Accurate and robust medical image segmentation is fundamental and crucial for enhancing the autonomy of computer-aided diagnosis and intervention systems. Medical data collection normally involves different scanners, protocols, and populations, making domain adaptation (DA) a highly demanding research field to alleviate model degradation in the deployment site. To preserve the model performance across multiple testing domains, this work proposes the Curriculum-based Augmented Fourier Domain Adaptation (Curri-AFDA) for robust medical image segmentation. In particular, our curriculum learning strategy is based on the causal relationship of a model under different levels of data shift in the deployment phase, where the higher the shift is, the harder to recognize the variance. Considering this, we progressively introduce more amplitude information from the target domain to the source domain in the frequency space during the curriculum-style training to smoothly schedule the semantic knowledge transfer in an easier-to-harder manner. Besides, we incorporate the training-time chained augmentation mixing to help expand the data distributions while preserving the domain-invariant semantics, which is beneficial for the acquired model to be more robust and generalize better to unseen domains. Extensive experiments on two segmentation tasks of Retina and Nuclei collected from multiple sites and scanners suggest that our proposed method yields superior adaptation and generalization performance. Meanwhile, our approach proves to be more robust under various corruption types and increasing severity levels. In addition, we show our method is also beneficial in the domain-adaptive classification task with skin lesion datasets. The code is available at https://github.com/lofrienger/Curri-AFDA.
comment: Work under review. First three authors contributed equally
☆ Semantic Segmentation on VSPW Dataset through Contrastive Loss and Multi-dataset Training Approach CVPR 2023
Video scene parsing incorporates temporal information, which can enhance the consistency and accuracy of predictions compared to image scene parsing. The added temporal dimension enables a more comprehensive understanding of the scene, leading to more reliable results. This paper presents the winning solution of the CVPR2023 workshop for video semantic segmentation, focusing on enhancing Spatial-Temporal correlations with contrastive loss. We also explore the influence of multi-dataset training by utilizing a label-mapping technique. And the final result is aggregating the output of the above two models. Our approach achieves 65.95% mIoU performance on the VSPW dataset, ranked 1st place on the VSPW challenge at CVPR 2023.
comment: 1st Place Solution for CVPR 2023 PVUW VSS Track
☆ Ada-TTA: Towards Adaptive High-Quality Text-to-Talking Avatar Synthesis
We are interested in a novel task, namely low-resource text-to-talking avatar. Given only a few-minute-long talking person video with the audio track as the training data and arbitrary texts as the driving input, we aim to synthesize high-quality talking portrait videos corresponding to the input text. This task has broad application prospects in the digital human industry but has not been technically achieved yet due to two challenges: (1) It is challenging to mimic the timbre from out-of-domain audio for a traditional multi-speaker Text-to-Speech system. (2) It is hard to render high-fidelity and lip-synchronized talking avatars with limited training data. In this paper, we introduce Adaptive Text-to-Talking Avatar (Ada-TTA), which (1) designs a generic zero-shot multi-speaker TTS model that well disentangles the text content, timbre, and prosody; and (2) embraces recent advances in neural rendering to achieve realistic audio-driven talking face video generation. With these designs, our method overcomes the aforementioned two challenges and achieves to generate identity-preserving speech and realistic talking person video. Experiments demonstrate that our method could synthesize realistic, identity-preserving, and audio-visual synchronized talking avatar videos.
comment: 6 pages, 3 figures
☆ Towards Adaptable and Interactive Image Captioning with Data Augmentation and Episodic Memory
Interactive machine learning (IML) is a beneficial learning paradigm in cases of limited data availability, as human feedback is incrementally integrated into the training process. In this paper, we present an IML pipeline for image captioning which allows us to incrementally adapt a pre-trained image captioning model to a new data distribution based on user input. In order to incorporate user input into the model, we explore the use of a combination of simple data augmentation methods to obtain larger data batches for each newly annotated data instance and implement continual learning methods to prevent catastrophic forgetting from repeated updates. For our experiments, we split a domain-specific image captioning dataset, namely VizWiz, into non-overlapping parts to simulate an incremental input flow for continually adapting the model to new data. We find that, while data augmentation worsens results, even when relatively small amounts of data are available, episodic memory is an effective strategy to retain knowledge from previously seen clusters.
☆ Instructive Feature Enhancement for Dichotomous Medical Image Segmentation MICCAI 2023
Deep neural networks have been widely applied in dichotomous medical image segmentation (DMIS) of many anatomical structures in several modalities, achieving promising performance. However, existing networks tend to struggle with task-specific, heavy and complex designs to improve accuracy. They made little instructions to which feature channels would be more beneficial for segmentation, and that may be why the performance and universality of these segmentation models are hindered. In this study, we propose an instructive feature enhancement approach, namely IFE, to adaptively select feature channels with rich texture cues and strong discriminability to enhance raw features based on local curvature or global information entropy criteria. Being plug-and-play and applicable for diverse DMIS tasks, IFE encourages the model to focus on texture-rich features which are especially important for the ambiguous and challenging boundary identification, simultaneously achieving simplicity, universality, and certain interpretability. To evaluate the proposed IFE, we constructed the first large-scale DMIS dataset Cosmos55k, which contains 55,023 images from 7 modalities and 26 anatomical structures. Extensive experiments show that IFE can improve the performance of classic segmentation networks across different anatomies and modalities with only slight modifications. Code is available at https://github.com/yezi-66/IFE
comment: Accepted by MICCAI 2023
☆ LegoNet: Alternating Model Blocks for Medical Image Segmentation
Since the emergence of convolutional neural networks (CNNs), and later vision transformers (ViTs), the common paradigm for model development has always been using a set of identical block types with varying parameters/hyper-parameters. To leverage the benefits of different architectural designs (e.g. CNNs and ViTs), we propose to alternate structurally different types of blocks to generate a new architecture, mimicking how Lego blocks can be assembled together. Using two CNN-based and one SwinViT-based blocks, we investigate three variations to the so-called LegoNet that applies the new concept of block alternation for the segmentation task in medical imaging. We also study a new clinical problem which has not been investigated before, namely the right internal mammary artery (RIMA) and perivascular space segmentation from computed tomography angiography (CTA) which has demonstrated a prognostic value to major cardiovascular outcomes. We compare the model performance against popular CNN and ViT architectures using two large datasets (e.g. achieving 0.749 dice similarity coefficient (DSC) on the larger dataset). We evaluate the performance of the model on three external testing cohorts as well, where an expert clinician made corrections to the model segmented results (DSC>0.90 for the three cohorts). To assess our proposed model for suitability in clinical use, we perform intra- and inter-observer variability analysis. Finally, we investigate a joint self-supervised learning approach to assess its impact on model performance. The code and the pretrained model weights will be available upon acceptance.
comment: 12 pages, 5 figures, 4 tables
☆ Efficient Anomaly Detection with Budget Annotation Using Semi-Supervised Residual Transformer
Anomaly Detection is challenging as usually only the normal samples are seen during training and the detector needs to discover anomalies on-the-fly. The recently proposed deep-learning-based approaches could somehow alleviate the problem but there is still a long way to go in obtaining an industrial-class anomaly detector for real-world applications. On the other hand, in some particular AD tasks, a few anomalous samples are labeled manually for achieving higher accuracy. However, this performance gain is at the cost of considerable annotation efforts, which can be intractable in many practical scenarios. In this work, the above two problems are addressed in a unified framework. Firstly, inspired by the success of the patch-matching-based AD algorithms, we train a sliding vision transformer over the residuals generated by a novel position-constrained patch-matching. Secondly, the conventional pixel-wise segmentation problem is cast into a block-wise classification problem. Thus the sliding transformer can attain even higher accuracy with much less annotation labor. Thirdly, to further reduce the labeling cost, we propose to label the anomalous regions using only bounding boxes. The unlabeled regions caused by the weak labels are effectively exploited using a highly-customized semi-supervised learning scheme equipped with two novel data augmentation methods. The proposed method outperforms all the state-of-the-art approaches using all the evaluation metrics in both the unsupervised and supervised scenarios. On the popular MVTec-AD dataset, our SemiREST algorithm obtains the Average Precision (AP) of 81.2% in the unsupervised condition and 84.4% AP for supervised anomaly detection. Surprisingly, with the bounding-box-based semi-supervisions, SemiREST still outperforms the SOTA methods with full supervision (83.8% AP) on MVTec-AD.
comment: 20 pages,6 figures
☆ SciCap+: A Knowledge Augmented Dataset to Study the Challenges of Scientific Figure Captioning AAAI23
In scholarly documents, figures provide a straightforward way of communicating scientific findings to readers. Automating figure caption generation helps move model understandings of scientific documents beyond text and will help authors write informative captions that facilitate communicating scientific findings. Unlike previous studies, we reframe scientific figure captioning as a knowledge-augmented image captioning task that models need to utilize knowledge embedded across modalities for caption generation. To this end, we extended the large-scale SciCap dataset~\cite{hsu-etal-2021-scicap-generating} to SciCap+ which includes mention-paragraphs (paragraphs mentioning figures) and OCR tokens. Then, we conduct experiments with the M4C-Captioner (a multimodal transformer-based model with a pointer network) as a baseline for our study. Our results indicate that mention-paragraphs serves as additional context knowledge, which significantly boosts the automatic standard image caption evaluation scores compared to the figure-only baselines. Human evaluations further reveal the challenges of generating figure captions that are informative to readers. The code and SciCap+ dataset will be publicly available at https://github.com/ZhishenYang/scientific_figure_captioning_dataset
comment: Published in SDU workshop at AAAI23
☆ Looking and Listening: Audio Guided Text Recognition
Text recognition in the wild is a long-standing problem in computer vision. Driven by end-to-end deep learning, recent studies suggest vision and language processing are effective for scene text recognition. Yet, solving edit errors such as add, delete, or replace is still the main challenge for existing approaches. In fact, the content of the text and its audio are naturally corresponding to each other, i.e., a single character error may result in a clear different pronunciation. In this paper, we propose the AudioOCR, a simple yet effective probabilistic audio decoder for mel spectrogram sequence prediction to guide the scene text recognition, which only participates in the training phase and brings no extra cost during the inference stage. The underlying principle of AudioOCR can be easily applied to the existing approaches. Experiments using 7 previous scene text recognition methods on 12 existing regular, irregular, and occluded benchmarks demonstrate our proposed method can bring consistent improvement. More importantly, through our experimentation, we show that AudioOCR possesses a generalizability that extends to more challenging scenarios, including recognizing non-English text, out-of-vocabulary words, and text with various accents. Code will be available at https://github.com/wenwenyu/AudioOCR.
☆ Putting Humans in the Image Captioning Loop
Image Captioning (IC) models can highly benefit from human feedback in the training process, especially in cases where data is limited. We present work-in-progress on adapting an IC system to integrate human feedback, with the goal to make it easily adaptable to user-specific data. Our approach builds on a base IC model pre-trained on the MS COCO dataset, which generates captions for unseen images. The user will then be able to offer feedback on the image and the generated/predicted caption, which will be augmented to create additional training instances for the adaptation of the model. The additional instances are integrated into the model using step-wise updates, and a sparse memory replay component is used to avoid catastrophic forgetting. We hope that this approach, while leading to improved results, will also result in customizable IC models.
☆ Benchmarking Robustness of AI-enabled Multi-sensor Fusion Systems: Challenges and Opportunities
Multi-Sensor Fusion (MSF) based perception systems have been the foundation in supporting many industrial applications and domains, such as self-driving cars, robotic arms, and unmanned aerial vehicles. Over the past few years, the fast progress in data-driven artificial intelligence (AI) has brought a fast-increasing trend to empower MSF systems by deep learning techniques to further improve performance, especially on intelligent systems and their perception systems. Although quite a few AI-enabled MSF perception systems and techniques have been proposed, up to the present, limited benchmarks that focus on MSF perception are publicly available. Given that many intelligent systems such as self-driving cars are operated in safety-critical contexts where perception systems play an important role, there comes an urgent need for a more in-depth understanding of the performance and reliability of these MSF systems. To bridge this gap, we initiate an early step in this direction and construct a public benchmark of AI-enabled MSF-based perception systems including three commonly adopted tasks (i.e., object detection, object tracking, and depth completion). Based on this, to comprehensively understand MSF systems' robustness and reliability, we design 14 common and realistic corruption patterns to synthesize large-scale corrupted datasets. We further perform a systematic evaluation of these systems through our large-scale evaluation. Our results reveal the vulnerability of the current AI-enabled MSF perception systems, calling for researchers and practitioners to take robustness and reliability into account when designing AI-enabled MSF.
comment: Accepted by ESEC/FSE 2023
☆ MetaGait: Learning to Learn an Omni Sample Adaptive Representation for Gait Recognition ECCV2022
Gait recognition, which aims at identifying individuals by their walking patterns, has recently drawn increasing research attention. However, gait recognition still suffers from the conflicts between the limited binary visual clues of the silhouette and numerous covariates with diverse scales, which brings challenges to the model's adaptiveness. In this paper, we address this conflict by developing a novel MetaGait that learns to learn an omni sample adaptive representation. Towards this goal, MetaGait injects meta-knowledge, which could guide the model to perceive sample-specific properties, into the calibration network of the attention mechanism to improve the adaptiveness from the omni-scale, omni-dimension, and omni-process perspectives. Specifically, we leverage the meta-knowledge across the entire process, where Meta Triple Attention and Meta Temporal Pooling are presented respectively to adaptively capture omni-scale dependency from spatial/channel/temporal dimensions simultaneously and to adaptively aggregate temporal information through integrating the merits of three complementary temporal aggregation methods. Extensive experiments demonstrate the state-of-the-art performance of the proposed MetaGait. On CASIA-B, we achieve rank-1 accuracy of 98.7%, 96.0%, and 89.3% under three conditions, respectively. On OU-MVLP, we achieve rank-1 accuracy of 92.4%.
comment: Accepted by ECCV2022
☆ DFormer: Diffusion-guided Transformer for Universal Image Segmentation
This paper introduces an approach, named DFormer, for universal image segmentation. The proposed DFormer views universal image segmentation task as a denoising process using a diffusion model. DFormer first adds various levels of Gaussian noise to ground-truth masks, and then learns a model to predict denoising masks from corrupted masks. Specifically, we take deep pixel-level features along with the noisy masks as inputs to generate mask features and attention masks, employing diffusion-based decoder to perform mask prediction gradually. At inference, our DFormer directly predicts the masks and corresponding categories from a set of randomly-generated masks. Extensive experiments reveal the merits of our proposed contributions on different image segmentation tasks: panoptic segmentation, instance segmentation, and semantic segmentation. Our DFormer outperforms the recent diffusion-based panoptic segmentation method Pix2Seq-D with a gain of 3.6% on MS COCO val2017 set. Further, DFormer achieves promising semantic segmentation performance outperforming the recent diffusion-based method by 2.2% on ADE20K val set. Our source code and models will be publicly on https://github.com/cp3wan/DFormer
☆ Revisiting the Trade-off between Accuracy and Robustness via Weight Distribution of Filters
Adversarial attacks have been proven to be potential threats to Deep Neural Networks (DNNs), and many methods are proposed to defend against adversarial attacks. However, while enhancing the robustness, the clean accuracy will decline to a certain extent, implying a trade-off existed between the accuracy and robustness. In this paper, we firstly empirically find an obvious distinction between standard and robust models in the filters' weight distribution of the same architecture, and then theoretically explain this phenomenon in terms of the gradient regularization, which shows this difference is an intrinsic property for DNNs, and thus a static network architecture is difficult to improve the accuracy and robustness at the same time. Secondly, based on this observation, we propose a sample-wise dynamic network architecture named Adversarial Weight-Varied Network (AW-Net), which focuses on dealing with clean and adversarial examples with a ``divide and rule" weight strategy. The AW-Net dynamically adjusts network's weights based on regulation signals generated by an adversarial detector, which is directly influenced by the input sample. Benefiting from the dynamic network architecture, clean and adversarial examples can be processed with different network weights, which provides the potentiality to enhance the accuracy and robustness simultaneously. A series of experiments demonstrate that our AW-Net is architecture-friendly to handle both clean and adversarial examples and can achieve better trade-off performance than state-of-the-art robust models.
☆ GaitGCI: Generative Counterfactual Intervention for Gait Recognition CVPR2023
Gait is one of the most promising biometrics that aims to identify pedestrians from their walking patterns. However, prevailing methods are susceptible to confounders, resulting in the networks hardly focusing on the regions that reflect effective walking patterns. To address this fundamental problem in gait recognition, we propose a Generative Counterfactual Intervention framework, dubbed GaitGCI, consisting of Counterfactual Intervention Learning (CIL) and Diversity-Constrained Dynamic Convolution (DCDC). CIL eliminates the impacts of confounders by maximizing the likelihood difference between factual/counterfactual attention while DCDC adaptively generates sample-wise factual/counterfactual attention to efficiently perceive the sample-wise properties. With matrix decomposition and diversity constraint, DCDC guarantees the model to be efficient and effective. Extensive experiments indicate that proposed GaitGCI: 1) could effectively focus on the discriminative and interpretable regions that reflect gait pattern; 2) is model-agnostic and could be plugged into existing models to improve performance with nearly no extra cost; 3) efficiently achieves state-of-the-art performance on arbitrary scenarios (in-the-lab and in-the-wild).
comment: Accepted by CVPR2023
☆ Change Diffusion: Change Detection Map Generation Based on Difference-Feature Guided DDPM
Deep learning (DL) approaches based on CNN-purely or Transformer networks have demonstrated promising results in bitemporal change detection (CD). However, their performance is limited by insufficient contextual information aggregation, as they struggle to fully capture the implicit contextual dependency relationships among feature maps at different levels. Additionally, researchers have utilized pre-trained denoising diffusion probabilistic models (DDPMs) for training lightweight CD classifiers. Nevertheless, training a DDPM to generate intricately detailed, multi-channel remote sensing images requires months of training time and a substantial volume of unlabeled remote sensing datasets, making it significantly more complex than generating a single-channel change map. To overcome these challenges, we propose a novel end-to-end DDPM-based model architecture called change-aware diffusion model (CADM), which can be trained using a limited annotated dataset quickly. Furthermore, we introduce dynamic difference conditional encoding to enhance step-wise regional attention in DDPM for bitemporal images in CD datasets. This method establishes state-adaptive conditions for each sampling step, emphasizing two main innovative points of our model: 1) its end-to-end nature and 2) difference conditional encoding. We evaluate CADM on four remote sensing CD tasks with different ground scenarios, including CDD, WHU, Levier, and GVLM. Experimental results demonstrate that CADM significantly outperforms state-of-the-art methods, indicating the generalization and effectiveness of the proposed model.
Prompting Large Language Models to Reformulate Queries for Moment Localization
The task of moment localization is to localize a temporal moment in an untrimmed video for a given natural language query. Since untrimmed video contains highly redundant contents, the quality of the query is crucial for accurately localizing moments, i.e., the query should provide precise information about the target moment so that the localization model can understand what to look for in the videos. However, the natural language queries in current datasets may not be easy to understand for existing models. For example, the Ego4D dataset uses question sentences as the query to describe relatively complex moments. While being natural and straightforward for humans, understanding such question sentences are challenging for mainstream moment localization models like 2D-TAN. Inspired by the recent success of large language models, especially their ability of understanding and generating complex natural language contents, in this extended abstract, we make early attempts at reformulating the moment queries into a set of instructions using large language models and making them more friendly to the localization models.
comment: 4 pages, 2 figures
☆ Diversifying Joint Vision-Language Tokenization Learning CVPR 2023
Building joint representations across images and text is an essential step for tasks such as Visual Question Answering and Video Question Answering. In this work, we find that the representations must not only jointly capture features from both modalities but should also be diverse for better generalization performance. To this end, we propose joint vision-language representation learning by diversifying the tokenization learning process, enabling tokens that are sufficiently disentangled from each other to be learned from both modalities. We observe that our approach outperforms the baseline models in a majority of settings and is competitive with state-of-the-art methods.
comment: Accepted to Transformers for Vision (T4V) workshop, CVPR 2023; 7 pages, 5 figures
☆ DreamSparse: Escaping from Plato's Cave with 2D Diffusion Model Given Sparse Views
Synthesizing novel view images from a few views is a challenging but practical problem. Existing methods often struggle with producing high-quality results or necessitate per-object optimization in such few-view settings due to the insufficient information provided. In this work, we explore leveraging the strong 2D priors in pre-trained diffusion models for synthesizing novel view images. 2D diffusion models, nevertheless, lack 3D awareness, leading to distorted image synthesis and compromising the identity. To address these problems, we propose DreamSparse, a framework that enables the frozen pre-trained diffusion model to generate geometry and identity-consistent novel view image. Specifically, DreamSparse incorporates a geometry module designed to capture 3D features from sparse views as a 3D prior. Subsequently, a spatial guidance model is introduced to convert these 3D feature maps into spatial information for the generative process. This information is then used to guide the pre-trained diffusion model, enabling it to generate geometrically consistent images without tuning it. Leveraging the strong image priors in the pre-trained diffusion models, DreamSparse is capable of synthesizing high-quality novel views for both object and scene-level images and generalising to open-set images. Experimental results demonstrate that our framework can effectively synthesize novel view images from sparse views and outperforms baselines in both trained and open-set category images. More results can be found on our project page: https://sites.google.com/view/dreamsparse-webpage.
☆ DVIS: Decoupled Video Instance Segmentation Framework
Video instance segmentation (VIS) is a critical task with diverse applications, including autonomous driving and video editing. Existing methods often underperform on complex and long videos in real world, primarily due to two factors. Firstly, offline methods are limited by the tightly-coupled modeling paradigm, which treats all frames equally and disregards the interdependencies between adjacent frames. Consequently, this leads to the introduction of excessive noise during long-term temporal alignment. Secondly, online methods suffer from inadequate utilization of temporal information. To tackle these challenges, we propose a decoupling strategy for VIS by dividing it into three independent sub-tasks: segmentation, tracking, and refinement. The efficacy of the decoupling strategy relies on two crucial elements: 1) attaining precise long-term alignment outcomes via frame-by-frame association during tracking, and 2) the effective utilization of temporal information predicated on the aforementioned accurate alignment outcomes during refinement. We introduce a novel referring tracker and temporal refiner to construct the \textbf{D}ecoupled \textbf{VIS} framework (\textbf{DVIS}). DVIS achieves new SOTA performance in both VIS and VPS, surpassing the current SOTA methods by 7.3 AP and 9.6 VPQ on the OVIS and VIPSeg datasets, which are the most challenging and realistic benchmarks. Moreover, thanks to the decoupling strategy, the referring tracker and temporal refiner are super light-weight (only 1.69\% of the segmenter FLOPs), allowing for efficient training and inference on a single GPU with 11G memory. The code is available at \href{https://github.com/zhang-tao-whu/DVIS}{https://github.com/zhang-tao-whu/DVIS}.
☆ Clinical-Inspired Cytological Whole Slide Image Screening with Just Slide-Level Labels
Cytology test is effective, non-invasive, convenient, and inexpensive for clinical cancer screening. ThinPrep, a commonly used liquid-based specimen, can be scanned to generate digital whole slide images (WSIs) for cytology testing. However, WSIs classification with gigapixel resolutions is highly resource-intensive, posing significant challenges for automated medical image analysis. In order to circumvent this computational impasse, existing methods emphasize learning features at the cell or patch level, typically requiring labor-intensive and detailed manual annotations, such as labels at the cell or patch level. Here we propose a novel automated Label-Efficient WSI Screening method, dubbed LESS, for cytology-based diagnosis with only slide-level labels. Firstly, in order to achieve label efficiency, we suggest employing variational positive-unlabeled (VPU) learning, enhancing patch-level feature learning using WSI-level labels. Subsequently, guided by the clinical approach of scrutinizing WSIs at varying fields of view and scales, we employ a cross-attention vision transformer (CrossViT) to fuse multi-scale patch-level data and execute WSI-level classification. We validate the proposed label-efficient method on a urine cytology WSI dataset encompassing 130 samples (13,000 patches) and FNAC 2019 dataset with 212 samples (21,200 patches). The experiment shows that the proposed LESS reaches 84.79%, 85.43%, 91.79% and 78.30% on a urine cytology WSI dataset, and 96.53%, 96.37%, 99.31%, 94.95% on FNAC 2019 dataset in terms of accuracy, AUC, sensitivity and specificity. It outperforms state-of-the-art methods and realizes automatic cytology-based bladder cancer screening.
comment: This paper was submitted to IEEE Transaction on Medical Imaging. It is under review
☆ Deep neural networks architectures from the perspective of manifold learning
Despite significant advances in the field of deep learning in ap-plications to various areas, an explanation of the learning pro-cess of neural network models remains an important open ques-tion. The purpose of this paper is a comprehensive comparison and description of neural network architectures in terms of ge-ometry and topology. We focus on the internal representation of neural networks and on the dynamics of changes in the topology and geometry of a data manifold on different layers. In this paper, we use the concepts of topological data analysis (TDA) and persistent homological fractal dimension. We present a wide range of experiments with various datasets and configurations of convolutional neural network (CNNs) architectures and Transformers in CV and NLP tasks. Our work is a contribution to the development of the important field of explainable and interpretable AI within the framework of geometrical deep learning.
comment: 11 pages, 12 figures, PRAI2023. arXiv admin note: substantial text overlap with arXiv:2204.08624
☆ SGAT4PASS: Spherical Geometry-Aware Transformer for PAnoramic Semantic Segmentation IJCAI 2023
As an important and challenging problem in computer vision, PAnoramic Semantic Segmentation (PASS) gives complete scene perception based on an ultra-wide angle of view. Usually, prevalent PASS methods with 2D panoramic image input focus on solving image distortions but lack consideration of the 3D properties of original $360^{\circ}$ data. Therefore, their performance will drop a lot when inputting panoramic images with the 3D disturbance. To be more robust to 3D disturbance, we propose our Spherical Geometry-Aware Transformer for PAnoramic Semantic Segmentation (SGAT4PASS), considering 3D spherical geometry knowledge. Specifically, a spherical geometry-aware framework is proposed for PASS. It includes three modules, i.e., spherical geometry-aware image projection, spherical deformable patch embedding, and a panorama-aware loss, which takes input images with 3D disturbance into account, adds a spherical geometry-aware constraint on the existing deformable patch embedding, and indicates the pixel density of original $360^{\circ}$ data, respectively. Experimental results on Stanford2D3D Panoramic datasets show that SGAT4PASS significantly improves performance and robustness, with approximately a 2% increase in mIoU, and when small 3D disturbances occur in the data, the stability of our performance is improved by an order of magnitude. Our code and supplementary material are available at https://github.com/TencentARC/SGAT4PASS.
comment: Accepted by IJCAI 2023
☆ G-CAME: Gaussian-Class Activation Mapping Explainer for Object Detectors
Nowadays, deep neural networks for object detection in images are very prevalent. However, due to the complexity of these networks, users find it hard to understand why these objects are detected by models. We proposed Gaussian Class Activation Mapping Explainer (G-CAME), which generates a saliency map as the explanation for object detection models. G-CAME can be considered a CAM-based method that uses the activation maps of selected layers combined with the Gaussian kernel to highlight the important regions in the image for the predicted box. Compared with other Region-based methods, G-CAME can transcend time constraints as it takes a very short time to explain an object. We also evaluated our method qualitatively and quantitatively with YOLOX on the MS-COCO 2017 dataset and guided to apply G-CAME into the two-stage Faster-RCNN model.
comment: 10 figures
☆ A Unified Framework to Super-Resolve Face Images of Varied Low Resolutions
The existing face image super-resolution (FSR) algorithms usually train a specific model for a specific low input resolution for optimal results. By contrast, we explore in this work a unified framework that is trained once and then used to super-resolve input face images of varied low resolutions. For that purpose, we propose a novel neural network architecture that is composed of three anchor auto-encoders, one feature weight regressor and a final image decoder. The three anchor auto-encoders are meant for optimal FSR for three pre-defined low input resolutions, or named anchor resolutions, respectively. An input face image of an arbitrary low resolution is firstly up-scaled to the target resolution by bi-cubic interpolation and then fed to the three auto-encoders in parallel. The three encoded anchor features are then fused with weights determined by the feature weight regressor. At last, the fused feature is sent to the final image decoder to derive the super-resolution result. As shown by experiments, the proposed algorithm achieves robust and state-of-the-art performance over a wide range of low input resolutions by a single framework. Code and models will be made available after the publication of this work.
☆ TextFormer: A Query-based End-to-End Text Spotter with Mixed Supervision
End-to-end text spotting is a vital computer vision task that aims to integrate scene text detection and recognition into a unified framework. Typical methods heavily rely on Region-of-Interest (RoI) operations to extract local features and complex post-processing steps to produce final predictions. To address these limitations, we propose TextFormer, a query-based end-to-end text spotter with Transformer architecture. Specifically, using query embedding per text instance, TextFormer builds upon an image encoder and a text decoder to learn a joint semantic understanding for multi-task modeling. It allows for mutual training and optimization of classification, segmentation, and recognition branches, resulting in deeper feature sharing without sacrificing flexibility or simplicity. Additionally, we design an Adaptive Global aGgregation (AGG) module to transfer global features into sequential features for reading arbitrarily-shaped texts, which overcomes the sub-optimization problem of RoI operations. Furthermore, potential corpus information is utilized from weak annotations to full labels through mixed supervision, further improving text detection and end-to-end text spotting results. Extensive experiments on various bilingual (i.e., English and Chinese) benchmarks demonstrate the superiority of our method. Especially on TDA-ReCTS dataset, TextFormer surpasses the state-of-the-art method in terms of 1-NED by 13.2%.
comment: MIR 2023, 15 pages
☆ Identifying Shared Decodable Concepts in the Human Brain Using Image-Language Foundation Models
We introduce a method that takes advantage of high-quality pretrained multimodal representations to explore fine-grained semantic networks in the human brain. Previous studies have documented evidence of functional localization in the brain, with different anatomical regions preferentially activating for different types of sensory input. Many such localized structures are known, including the fusiform face area and parahippocampal place area. This raises the question of whether additional brain regions (or conjunctions of brain regions) are also specialized for other important semantic concepts. To identify such brain regions, we developed a data-driven approach to uncover visual concepts that are decodable from a massive functional magnetic resonance imaging (fMRI) dataset. Our analysis is broadly split into three sections. First, a fully connected neural network is trained to map brain responses to the outputs of an image-language foundation model, CLIP (Radford et al., 2021). Subsequently, a contrastive-learning dimensionality reduction method reveals the brain-decodable components of CLIP space. In the final section of our analysis, we localize shared decodable concepts in the brain using a voxel-masking optimization method to produce a shared decodable concept (SDC) space. The accuracy of our procedure is validated by comparing it to previous localization experiments that identify regions for faces, bodies, and places. In addition to these concepts, whose corresponding brain regions were already known, we localize novel concept representations which are shared across participants to other areas of the human brain. We also demonstrate how this method can be used to inspect fine-grained semantic networks for individual participants. We envisage that this extensible method can also be adapted to explore other questions at the intersection of AI and neuroscience.
comment: Under review
☆ PGformer: Proxy-Bridged Game Transformer for Multi-Person Extremely Interactive Motion Prediction
Multi-person motion prediction is a challenging task, especially for real-world scenarios of densely interacted persons. Most previous works have been devoted to studying the case of weak interactions (e.g., hand-shaking), which typically forecast each human pose in isolation. In this paper, we focus on motion prediction for multiple persons with extreme collaborations and attempt to explore the relationships between the highly interactive persons' motion trajectories. Specifically, a novel cross-query attention (XQA) module is proposed to bilaterally learn the cross-dependencies between the two pose sequences tailored for this situation. Additionally, we introduce and build a proxy entity to bridge the involved persons, which cooperates with our proposed XQA module and subtly controls the bidirectional information flows, acting as a motion intermediary. We then adapt these designs to a Transformer-based architecture and devise a simple yet effective end-to-end framework called proxy-bridged game Transformer (PGformer) for multi-person interactive motion prediction. The effectiveness of our method has been evaluated on the challenging ExPI dataset, which involves highly interactive actions. We show that our PGformer consistently outperforms the state-of-the-art methods in both short- and long-term predictions by a large margin. Besides, our approach can also be compatible with the weakly interacted CMU-Mocap and MuPoTS-3D datasets and achieve encouraging results. Our code will become publicly available upon acceptance.
☆ CiT-Net: Convolutional Neural Networks Hand in Hand with Vision Transformers for Medical Image Segmentation
The hybrid architecture of convolutional neural networks (CNNs) and Transformer are very popular for medical image segmentation. However, it suffers from two challenges. First, although a CNNs branch can capture the local image features using vanilla convolution, it cannot achieve adaptive feature learning. Second, although a Transformer branch can capture the global features, it ignores the channel and cross-dimensional self-attention, resulting in a low segmentation accuracy on complex-content images. To address these challenges, we propose a novel hybrid architecture of convolutional neural networks hand in hand with vision Transformers (CiT-Net) for medical image segmentation. Our network has two advantages. First, we design a dynamic deformable convolution and apply it to the CNNs branch, which overcomes the weak feature extraction ability due to fixed-size convolution kernels and the stiff design of sharing kernel parameters among different inputs. Second, we design a shifted-window adaptive complementary attention module and a compact convolutional projection. We apply them to the Transformer branch to learn the cross-dimensional long-term dependency for medical images. Experimental results show that our CiT-Net provides better medical image segmentation results than popular SOTA methods. Besides, our CiT-Net requires lower parameters and less computational costs and does not rely on pre-training. The code is publicly available at https://github.com/SR0920/CiT-Net.
☆ Learning Representations on the Unit Sphere: Application to Online Continual Learning
We use the maximum a posteriori estimation principle for learning representations distributed on the unit sphere. We derive loss functions for the von Mises-Fisher distribution and the angular Gaussian distribution, both designed for modeling symmetric directional data. A noteworthy feature of our approach is that the learned representations are pushed toward fixed directions, allowing for a learning strategy that is resilient to data drift. This makes it suitable for online continual learning, which is the problem of training neural networks on a continuous data stream, where multiple classification tasks are presented sequentially so that data from past tasks are no longer accessible, and data from the current task can be seen only once. To address this challenging scenario, we propose a memory-based representation learning technique equipped with our new loss functions. Our approach does not require negative data or knowledge of task boundaries and performs well with smaller batch sizes while being computationally efficient. We demonstrate with extensive experiments that the proposed method outperforms the current state-of-the-art methods on both standard evaluation scenarios and realistic scenarios with blurry task boundaries. For reproducibility, we use the same training pipeline for every compared method and share the code at https://t.ly/SQTj.
comment: 16 pages, 4 figures, under review
☆ BatchSampler: Sampling Mini-Batches for Contrastive Learning in Vision, Language, and Graphs
In-Batch contrastive learning is a state-of-the-art self-supervised method that brings semantically-similar instances close while pushing dissimilar instances apart within a mini-batch. Its key to success is the negative sharing strategy, in which every instance serves as a negative for the others within the mini-batch. Recent studies aim to improve performance by sampling hard negatives \textit{within the current mini-batch}, whose quality is bounded by the mini-batch itself. In this work, we propose to improve contrastive learning by sampling mini-batches from the input data. We present BatchSampler\footnote{The code is available at \url{https://github.com/THUDM/BatchSampler}} to sample mini-batches of hard-to-distinguish (i.e., hard and true negatives to each other) instances. To make each mini-batch have fewer false negatives, we design the proximity graph of randomly-selected instances. To form the mini-batch, we leverage random walk with restart on the proximity graph to help sample hard-to-distinguish instances. BatchSampler is a simple and general technique that can be directly plugged into existing contrastive learning models in vision, language, and graphs. Extensive experiments on datasets of three modalities show that BatchSampler can consistently improve the performance of powerful contrastive models, as shown by significant improvements of SimCLR on ImageNet-100, SimCSE on STS (language), and GraphCL and MVGRL on graph datasets.
comment: 17 pages, 16 figures
☆ A Robust Likelihood Model for Novelty Detection CVPR
Current approaches to novelty or anomaly detection are based on deep neural networks. Despite their effectiveness, neural networks are also vulnerable to imperceptible deformations of the input data. This is a serious issue in critical applications, or when data alterations are generated by an adversarial attack. While this is a known problem that has been studied in recent years for the case of supervised learning, the case of novelty detection has received very limited attention. Indeed, in this latter setting the learning is typically unsupervised because outlier data is not available during training, and new approaches for this case need to be investigated. We propose a new prior that aims at learning a robust likelihood for the novelty test, as a defense against attacks. We also integrate the same prior with a state-of-the-art novelty detection approach. Because of the geometric properties of that approach, the resulting robust training is computationally very efficient. An initial evaluation of the method indicates that it is effective at improving performance with respect to the standard models in the absence and presence of attacks.
comment: CVPR Workshop on Computer Vision in the Wild, 2023
☆ Student Classroom Behavior Detection based on Improved YOLOv7
Accurately detecting student behavior in classroom videos can aid in analyzing their classroom performance and improving teaching effectiveness. However, the current accuracy rate in behavior detection is low. To address this challenge, we propose the Student Classroom Behavior Detection method, based on improved YOLOv7. First, we created the Student Classroom Behavior dataset (SCB-Dataset), which includes 18.4k labels and 4.2k images, covering three behaviors: hand raising, reading, and writing. To improve detection accuracy in crowded scenes, we integrated the biformer attention module and Wise-IoU into the YOLOv7 network. Finally, experiments were conducted on the SCB-Dataset, and the model achieved an mAP@0.5 of 79%, resulting in a 1.8% improvement over previous results. The SCB-Dataset and code are available for download at: https://github.com/Whiffe/SCB-dataset.
comment: arXiv admin note: text overlap with arXiv:2305.07825
☆ Active Sparse Conversations for Improved Audio-Visual Embodied Navigation
Efficient navigation towards an audio-goal necessitates an embodied agent to not only possess the ability to use audio-visual cues effectively, but also be equipped to actively (but occasionally) seek human/oracle assistance without sacrificing autonomy, e.g., when it is uncertain of where to navigate towards locating a noisy or sporadic audio goal. To this end, we present CAVEN -- a conversational audio-visual embodied navigation agent that is capable of posing navigation questions to a human/oracle and processing the oracle responses; both in free-form natural language. At the core of CAVEN is a multimodal hierarchical reinforcement learning (RL) setup that is equipped with a high-level policy that is trained to choose from one of three low-level policies (at every step), namely: (i) to navigate using audio-visual cues, or (ii) to frame a question to the oracle and receive a short or detailed response, or (iii) ask generic questions (when unsure of what to ask) and receive instructions. Key to generating the agent's questions is our novel TrajectoryNet that forecasts the most likely next steps to the goal and a QuestionNet that uses these steps to produce a question. All the policies are learned end-to-end via the RL setup, with penalties to enforce sparsity in receiving navigation instructions from the oracle. To evaluate the performance of CAVEN, we present extensive experiments on the SoundSpaces framework for the task of semantic audio-visual navigation. Our results show that CAVEN achieves upto 12% gain in performance over competing methods, especially in localizing new sound sources, even in the presence of auditory distractions.
☆ Quantitative Analysis of Primary Attribution Explainable Artificial Intelligence Methods for Remote Sensing Image Classification
We present a comprehensive analysis of quantitatively evaluating explainable artificial intelligence (XAI) techniques for remote sensing image classification. Our approach leverages state-of-the-art machine learning approaches to perform remote sensing image classification across multiple modalities. We investigate the results of the models qualitatively through XAI methods. Additionally, we compare the XAI methods quantitatively through various categories of desired properties. Through our analysis, we offer insights and recommendations for selecting the most appropriate XAI method(s) to gain a deeper understanding of the models' decision-making processes. The code for this work is publicly available.
comment: 4 pages, 3 figures, Accepted to 2023 IGARSS Community-Contributed Sessions - Opening the Black Box: Explainable AI/ML in Remote Sensing Analysis
☆ BokehOrNot: Transforming Bokeh Effect with Image Transformer and Lens Metadata Embedding
Bokeh effect is an optical phenomenon that offers a pleasant visual experience, typically generated by high-end cameras with wide aperture lenses. The task of bokeh effect transformation aims to produce a desired effect in one set of lenses and apertures based on another combination. Current models are limited in their ability to render a specific set of bokeh effects, primarily transformations from sharp to blur. In this paper, we propose a novel universal method for embedding lens metadata into the model and introducing a loss calculation method using alpha masks from the newly released Bokeh Effect Transformation Dataset(BETD) [3]. Based on the above techniques, we propose the BokehOrNot model, which is capable of producing both blur-to-sharp and sharp-to-blur bokeh effect with various combinations of lenses and aperture sizes. Our proposed model outperforms current leading bokeh rendering and image restoration models and renders visually natural bokeh effects. Our code is available at: https://github.com/indicator0/bokehornot.
☆ Energy-Based Models for Cross-Modal Localization using Convolutional Transformers ICRA 2023
We present a novel framework using Energy-Based Models (EBMs) for localizing a ground vehicle mounted with a range sensor against satellite imagery in the absence of GPS. Lidar sensors have become ubiquitous on autonomous vehicles for describing its surrounding environment. Map priors are typically built using the same sensor modality for localization purposes. However, these map building endeavors using range sensors are often expensive and time-consuming. Alternatively, we leverage the use of satellite images as map priors, which are widely available, easily accessible, and provide comprehensive coverage. We propose a method using convolutional transformers that performs accurate metric-level localization in a cross-modal manner, which is challenging due to the drastic difference in appearance between the sparse range sensor readings and the rich satellite imagery. We train our model end-to-end and demonstrate our approach achieving higher accuracy than the state-of-the-art on KITTI, Pandaset, and a custom dataset.
comment: ICRA 2023
☆ A Quality Aware Sample-to-Sample Comparison for Face Recognition WACV23
Currently available face datasets mainly consist of a large number of high-quality and a small number of low-quality samples. As a result, a Face Recognition (FR) network fails to learn the distribution of low-quality samples since they are less frequent during training (underrepresented). Moreover, current state-of-the-art FR training paradigms are based on the sample-to-center comparison (i.e., Softmax-based classifier), which results in a lack of uniformity between train and test metrics. This work integrates a quality-aware learning process at the sample level into the classification training paradigm (QAFace). In this regard, Softmax centers are adaptively guided to pay more attention to low-quality samples by using a quality-aware function. Accordingly, QAFace adds a quality-based adjustment to the updating procedure of the Softmax-based classifier to improve the performance on the underrepresented low-quality samples. Our method adaptively finds and assigns more attention to the recognizable low-quality samples in the training datasets. In addition, QAFace ignores the unrecognizable low-quality samples using the feature magnitude as a proxy for quality. As a result, QAFace prevents class centers from getting distracted from the optimal direction. The proposed method is superior to the state-of-the-art algorithms in extensive experimental results on the CFP-FP, LFW, CPLFW, CALFW, AgeDB, IJB-B, and IJB-C datasets.
comment: IEEE/CVF Winter Conference on Applications of Computer Vision (WACV23)
☆ Real-Time Online Unsupervised Domain Adaptation for Real-World Person Re-identification
Following the popularity of Unsupervised Domain Adaptation (UDA) in person re-identification, the recently proposed setting of Online Unsupervised Domain Adaptation (OUDA) attempts to bridge the gap towards practical applications by introducing a consideration of streaming data. However, this still falls short of truly representing real-world applications. This paper defines the setting of Real-world Real-time Online Unsupervised Domain Adaptation (R$^2$OUDA) for Person Re-identification. The R$^2$OUDA setting sets the stage for true real-world real-time OUDA, bringing to light four major limitations found in real-world applications that are often neglected in current research: system generated person images, subset distribution selection, time-based data stream segmentation, and a segment-based time constraint. To address all aspects of this new R$^2$OUDA setting, this paper further proposes Real-World Real-Time Online Streaming Mutual Mean-Teaching (R$^2$MMT), a novel multi-camera system for real-world person re-identification. Taking a popular person re-identification dataset, R$^2$MMT was used to construct over 100 data subsets and train more than 3000 models, exploring the breadth of the R$^2$OUDA setting to understand the training time and accuracy trade-offs and limitations for real-world applications. R$^2$MMT, a real-world system able to respect the strict constraints of the proposed R$^2$OUDA setting, achieves accuracies within 0.1% of comparable OUDA methods that cannot be applied directly to real-world applications.
☆ Learn the Force We Can: Multi-Object Video Generation from Pixel-Level Interactions
We propose a novel unsupervised method to autoregressively generate videos from a single frame and a sparse motion input. Our trained model can generate realistic object-to-object interactions and separate the dynamics and the extents of multiple objects despite only observing them under correlated motion activities. Key components in our method are the randomized conditioning scheme, the encoding of the input motion control, and the randomized and sparse sampling to break correlations. Our model, which we call YODA, has the ability to move objects without physically touching them. We show both qualitatively and quantitatively that YODA accurately follows the user control, while yielding a video quality that is on par with or better than state of the art video generation prior work on several datasets. For videos, visit our project website https://araachie.github.io/yoda.
comment: Project website: https://araachie.github.io/yoda
☆ Unsupervised Iterative U-Net with an Internal Guidance Layer for Vertebrae Contrast Enhancement in Chest X-Ray Images
X-ray imaging is a fundamental clinical tool for screening and diagnosing various diseases. However, the spatial resolution of radiographs is often limited, making it challenging to diagnose small image details and leading to difficulties in identifying vertebrae anomalies at an early stage in chest radiographs. To address this limitation, we propose a novel and robust approach to significantly improve the quality of X-ray images by iteratively training a deep neural network. Our framework includes an embedded internal guidance layer that enhances the fine structures of spinal vertebrae in chest X-ray images through fully unsupervised training, utilizing an iterative procedure that employs the same network architecture in each enhancement phase. Additionally, we have designed an optimized loss function that accurately identifies object boundaries and enhances spinal features, thereby further enhancing the quality of the images. Experimental results demonstrate that our proposed method surpasses existing detail enhancement methods in terms of BRISQUE scores, and is comparable in terms of LPC-SI. Furthermore, our approach exhibits superior performance in restoring hidden fine structures, as evidenced by our qualitative results. This innovative approach has the potential to significantly enhance the diagnostic accuracy and early detection of diseases, making it a promising advancement in X-ray imaging technology.
comment: 10 pages, 11 figures, submitted to Transactions on Medical Imaging
☆ Recognition of Handwritten Japanese Characters Using Ensemble of Convolutional Neural Networks
The Japanese writing system is complex, with three character types of Hiragana, Katakana, and Kanji. Kanji consists of thousands of unique characters, further adding to the complexity of character identification and literature understanding. Being able to translate handwritten Japanese characters into digital text is useful for data analysis, translation, learning and cultural preservation. In this study, a machine learning approach to analyzing and recognizing handwritten Japanese characters (Kanji) is proposed. The study used an ensemble of three convolutional neural networks (CNNs) for recognizing handwritten Kanji characters and utilized four datasets of MNIST, K-MNIST, Kuzushiji-49 (K49) and the top 150 represented classes in the Kuzushiji-Kanji (K-Kanji) dataset for its performance evaluation. The results indicate feasibility of using proposed CNN-ensemble architecture for recognizing handwritten characters, achieving 99.4%, 96.4%, 95.0% and 96.4% classification accuracy on MNIST, K-MNIS, K49, and K-Kanji datasets respectively.
☆ Accurate Fine-Grained Segmentation of Human Anatomy in Radiographs via Volumetric Pseudo-Labeling
Purpose: Interpreting chest radiographs (CXR) remains challenging due to the ambiguity of overlapping structures such as the lungs, heart, and bones. To address this issue, we propose a novel method for extracting fine-grained anatomical structures in CXR using pseudo-labeling of three-dimensional computed tomography (CT) scans. Methods: We created a large-scale dataset of 10,021 thoracic CTs with 157 labels and applied an ensemble of 3D anatomy segmentation models to extract anatomical pseudo-labels. These labels were projected onto a two-dimensional plane, similar to the CXR, allowing the training of detailed semantic segmentation models for CXR without any manual annotation effort. Results: Our resulting segmentation models demonstrated remarkable performance on CXR, with a high average model-annotator agreement between two radiologists with mIoU scores of 0.93 and 0.85 for frontal and lateral anatomy, while inter-annotator agreement remained at 0.95 and 0.83 mIoU. Our anatomical segmentations allowed for the accurate extraction of relevant explainable medical features such as the cardio-thoracic-ratio. Conclusion: Our method of volumetric pseudo-labeling paired with CT projection offers a promising approach for detailed anatomical segmentation of CXR with a high agreement with human annotators. This technique may have important clinical implications, particularly in the analysis of various thoracic pathologies.
comment: 28 pages, 1 table, 10 figures
☆ Q: How to Specialize Large Vision-Language Models to Data-Scarce VQA Tasks? A: Self-Train on Unlabeled Images! CVPR 2023
Finetuning a large vision language model (VLM) on a target dataset after large scale pretraining is a dominant paradigm in visual question answering (VQA). Datasets for specialized tasks such as knowledge-based VQA or VQA in non natural-image domains are orders of magnitude smaller than those for general-purpose VQA. While collecting additional labels for specialized tasks or domains can be challenging, unlabeled images are often available. We introduce SelTDA (Self-Taught Data Augmentation), a strategy for finetuning large VLMs on small-scale VQA datasets. SelTDA uses the VLM and target dataset to build a teacher model that can generate question-answer pseudolabels directly conditioned on an image alone, allowing us to pseudolabel unlabeled images. SelTDA then finetunes the initial VLM on the original dataset augmented with freshly pseudolabeled images. We describe a series of experiments showing that our self-taught data augmentation increases robustness to adversarially searched questions, counterfactual examples and rephrasings, improves domain generalization, and results in greater retention of numerical reasoning skills. The proposed strategy requires no additional annotations or architectural modifications, and is compatible with any modern encoder-decoder multimodal transformer. Code available at https://github.com/codezakh/SelTDA.
comment: CVPR 2023
♻ ☆ blob loss: instance imbalance aware loss functions for semantic segmentation
Deep convolutional neural networks (CNN) have proven to be remarkably effective in semantic segmentation tasks. Most popular loss functions were introduced targeting improved volumetric scores, such as the Dice coefficient (DSC). By design, DSC can tackle class imbalance, however, it does not recognize instance imbalance within a class. As a result, a large foreground instance can dominate minor instances and still produce a satisfactory DSC. Nevertheless, detecting tiny instances is crucial for many applications, such as disease monitoring. For example, it is imperative to locate and surveil small-scale lesions in the follow-up of multiple sclerosis patients. We propose a novel family of loss functions, \emph{blob loss}, primarily aimed at maximizing instance-level detection metrics, such as F1 score and sensitivity. \emph{Blob loss} is designed for semantic segmentation problems where detecting multiple instances matters. We extensively evaluate a DSC-based \emph{blob loss} in five complex 3D semantic segmentation tasks featuring pronounced instance heterogeneity in terms of texture and morphology. Compared to soft Dice loss, we achieve 5% improvement for MS lesions, 3% improvement for liver tumor, and an average 2% improvement for microscopy segmentation tasks considering F1 score.
comment: 23 pages, 7 figures // corrected one mistake where it said beta instead of alpha in the text
♻ ☆ Concept-based Explanations for Out-Of-Distribution Detectors ICML'23
Out-of-distribution (OOD) detection plays a crucial role in ensuring the safe deployment of deep neural network (DNN) classifiers. While a myriad of methods have focused on improving the performance of OOD detectors, a critical gap remains in interpreting their decisions. We help bridge this gap by providing explanations for OOD detectors based on learned high-level concepts. We first propose two new metrics for assessing the effectiveness of a particular set of concepts for explaining OOD detectors: 1) detection completeness, which quantifies the sufficiency of concepts for explaining an OOD-detector's decisions, and 2) concept separability, which captures the distributional separation between in-distribution and OOD data in the concept space. Based on these metrics, we propose an unsupervised framework for learning a set of concepts that satisfy the desired properties of high detection completeness and concept separability, and demonstrate its effectiveness in providing concept-based explanations for diverse off-the-shelf OOD detectors. We also show how to identify prominent concepts contributing to the detection results, and provide further reasoning about their decisions.
comment: Paper published at International Conference on Machine Learning (ICML'23)
♻ ☆ Bridging the Gap: Enhancing the Utility of Synthetic Data via Post-Processing Techniques
Acquiring and annotating suitable datasets for training deep learning models is challenging. This often results in tedious and time-consuming efforts that can hinder research progress. However, generative models have emerged as a promising solution for generating synthetic datasets that can replace or augment real-world data. Despite this, the effectiveness of synthetic data is limited by their inability to fully capture the complexity and diversity of real-world data. To address this issue, we explore the use of Generative Adversarial Networks to generate synthetic datasets for training classifiers that are subsequently evaluated on real-world images. To improve the quality and diversity of the synthetic dataset, we propose three novel post-processing techniques: Dynamic Sample Filtering, Dynamic Dataset Recycle, and Expansion Trick. In addition, we introduce a pipeline called Gap Filler (GaFi), which applies these techniques in an optimal and coordinated manner to maximise classification accuracy on real-world data. Our experiments show that GaFi effectively reduces the gap with real-accuracy scores to an error of 2.03%, 1.78%, and 3.99% on the Fashion-MNIST, CIFAR-10, and CIFAR-100 datasets, respectively. These results represent a new state of the art in Classification Accuracy Score and highlight the effectiveness of post-processing techniques in improving the quality of synthetic datasets.
♻ ☆ Prediction of Post-Operative Renal and Pulmonary Complications Using Transformers
Postoperative complications pose a significant challenge in the healthcare industry, resulting in elevated healthcare expenses and prolonged hospital stays, and in rare instances, patient mortality. To improve patient outcomes and reduce healthcare costs, healthcare providers rely on various perioperative risk scores to guide clinical decisions and prioritize care. In recent years, machine learning techniques have shown promise in predicting postoperative complications and fatality, with deep learning models achieving remarkable success in healthcare applications. However, research on the application of deep learning models to intra-operative anesthesia management data is limited. In this paper, we evaluate the performance of transformer-based models in predicting postoperative acute renal failure, postoperative pulmonary complications, and postoperative in-hospital mortality. We compare our method's performance with state-of-the-art tabular data prediction models, including gradient boosting trees and sequential attention models, on a clinical dataset. Our results demonstrate that transformer-based models can achieve superior performance in predicting postoperative complications and outperform traditional machine learning models. This work highlights the potential of deep learning techniques, specifically transformer-based models, in revolutionizing the healthcare industry's approach to postoperative care.
♻ ☆ Survey on LiDAR Perception in Adverse Weather Conditions
Autonomous vehicles rely on a variety of sensors to gather information about their surrounding. The vehicle's behavior is planned based on the environment perception, making its reliability crucial for safety reasons. The active LiDAR sensor is able to create an accurate 3D representation of a scene, making it a valuable addition for environment perception for autonomous vehicles. Due to light scattering and occlusion, the LiDAR's performance change under adverse weather conditions like fog, snow or rain. This limitation recently fostered a large body of research on approaches to alleviate the decrease in perception performance. In this survey, we gathered, analyzed, and discussed different aspects on dealing with adverse weather conditions in LiDAR-based environment perception. We address topics such as the availability of appropriate data, raw point cloud processing and denoising, robust perception algorithms and sensor fusion to mitigate adverse weather induced shortcomings. We furthermore identify the most pressing gaps in the current literature and pinpoint promising research directions.
comment: published at IEEE IV 2023
♻ ☆ Domain Generalization for Mammographic Image Analysis via Contrastive Learning
The deep learning technique has been shown to be effective in addressing several image analysis tasks within the computer-aided diagnosis scheme for mammography. The training of an efficacious deep learning model requires large amounts of data with sufficient diversity in terms of image style and quality. In particular, the diversity of image styles may be primarily attributed to the vendor factor. However, the collection of mammograms from large and diverse vendors is very expensive and sometimes impractical. Motivatedly, a novel contrastive learning method is developed to equip the deep learning models with better generalization capability. Specifically, the multi-style and multi-view unsupervised self-learning scheme is carried out to seek robust feature embedding against various vendor styles as a pre-trained model. Afterward, the pre-trained network is further fine-tuned to the downstream tasks, e.g., mass detection, matching, BI-RADS rating, and breast density classification. The proposed method has been extensively and rigorously evaluated with mammograms from various vendor-style domains and several public datasets. The experimental results suggest that the proposed domain generalization method can effectively improve the performance of four mammographic image tasks on data from either seen or unseen domains and outperform many state-of-the-art (SOTA) generalization methods.
comment: arXiv admin note: text overlap with arXiv:2111.10827
♻ ☆ Near-field Perception for Low-Speed Vehicle Automation using Surround-view Fisheye Cameras
Cameras are the primary sensor in automated driving systems. They provide high information density and are optimal for detecting road infrastructure cues laid out for human vision. Surround-view camera systems typically comprise of four fisheye cameras with 190{\deg}+ field of view covering the entire 360{\deg} around the vehicle focused on near-field sensing. They are the principal sensors for low-speed, high accuracy, and close-range sensing applications, such as automated parking, traffic jam assistance, and low-speed emergency braking. In this work, we provide a detailed survey of such vision systems, setting up the survey in the context of an architecture that can be decomposed into four modular components namely Recognition, Reconstruction, Relocalization, and Reorganization. We jointly call this the 4R Architecture. We discuss how each component accomplishes a specific aspect and provide a positional argument that they can be synergized to form a complete perception system for low-speed automation. We support this argument by presenting results from previous works and by presenting architecture proposals for such a system. Qualitative results are presented in the video at https://youtu.be/ae8bCOF77uY.
comment: Accepted for publication at IEEE Transactions on Intelligent Transportation Systems
♻ ☆ OmniDet: Surround View Cameras based Multi-task Visual Perception Network for Autonomous Driving ICRA 2021
Surround View fisheye cameras are commonly deployed in automated driving for 360\deg{} near-field sensing around the vehicle. This work presents a multi-task visual perception network on unrectified fisheye images to enable the vehicle to sense its surrounding environment. It consists of six primary tasks necessary for an autonomous driving system: depth estimation, visual odometry, semantic segmentation, motion segmentation, object detection, and lens soiling detection. We demonstrate that the jointly trained model performs better than the respective single task versions. Our multi-task model has a shared encoder providing a significant computational advantage and has synergized decoders where tasks support each other. We propose a novel camera geometry based adaptation mechanism to encode the fisheye distortion model both at training and inference. This was crucial to enable training on the WoodScape dataset, comprised of data from different parts of the world collected by 12 different cameras mounted on three different cars with different intrinsics and viewpoints. Given that bounding boxes is not a good representation for distorted fisheye images, we also extend object detection to use a polygon with non-uniformly sampled vertices. We additionally evaluate our model on standard automotive datasets, namely KITTI and Cityscapes. We obtain the state-of-the-art results on KITTI for depth estimation and pose estimation tasks and competitive performance on the other tasks. We perform extensive ablation studies on various architecture choices and task weighting methodologies. A short video at https://youtu.be/xbSjZ5OfPes provides qualitative results.
comment: Best Robot Vision paper award finalist (top 4). Camera ready version accepted for RA-L and ICRA 2021 publication
♻ ☆ UnRectDepthNet: Self-Supervised Monocular Depth Estimation using a Generic Framework for Handling Common Camera Distortion Models IROS 2020
In classical computer vision, rectification is an integral part of multi-view depth estimation. It typically includes epipolar rectification and lens distortion correction. This process simplifies the depth estimation significantly, and thus it has been adopted in CNN approaches. However, rectification has several side effects, including a reduced field of view (FOV), resampling distortion, and sensitivity to calibration errors. The effects are particularly pronounced in case of significant distortion (e.g., wide-angle fisheye cameras). In this paper, we propose a generic scale-aware self-supervised pipeline for estimating depth, euclidean distance, and visual odometry from unrectified monocular videos. We demonstrate a similar level of precision on the unrectified KITTI dataset with barrel distortion comparable to the rectified KITTI dataset. The intuition being that the rectification step can be implicitly absorbed within the CNN model, which learns the distortion model without increasing complexity. Our approach does not suffer from a reduced field of view and avoids computational costs for rectification at inference time. To further illustrate the general applicability of the proposed framework, we apply it to wide-angle fisheye cameras with 190$^\circ$ horizontal field of view. The training framework UnRectDepthNet takes in the camera distortion model as an argument and adapts projection and unprojection functions accordingly. The proposed algorithm is evaluated further on the KITTI rectified dataset, and we achieve state-of-the-art results that improve upon our previous work FisheyeDistanceNet. Qualitative results on a distorted test scene video sequence indicate excellent performance https://youtu.be/K6pbx3bU4Ss.
comment: Minor fixes added after IROS 2020 Camera ready submission. IROS 2020 presentation video - https://www.youtube.com/watch?v=3Br2KSWZRrY
♻ ☆ Provable Dynamic Fusion for Low-Quality Multimodal Data ICML 2023
The inherent challenge of multimodal fusion is to precisely capture the cross-modal correlation and flexibly conduct cross-modal interaction. To fully release the value of each modality and mitigate the influence of low-quality multimodal data, dynamic multimodal fusion emerges as a promising learning paradigm. Despite its widespread use, theoretical justifications in this field are still notably lacking. Can we design a provably robust multimodal fusion method? This paper provides theoretical understandings to answer this question under a most popular multimodal fusion framework from the generalization perspective. We proceed to reveal that several uncertainty estimation solutions are naturally available to achieve robust multimodal fusion. Then a novel multimodal fusion framework termed Quality-aware Multimodal Fusion (QMF) is proposed, which can improve the performance in terms of classification accuracy and model robustness. Extensive experimental results on multiple benchmarks can support our findings.
comment: Accepted by ICML 2023
♻ ☆ Learning to predict 3D rotational dynamics from images of a rigid body with unknown mass distribution
In many real-world settings, image observations of freely rotating 3D rigid bodies, may be available when low-dimensional measurements are not. However, the high-dimensionality of image data precludes the use of classical estimation techniques to learn the dynamics. The usefulness of standard deep learning methods is also limited because an image of a rigid body reveals nothing about the distribution of mass inside the body, which, together with initial angular velocity, is what determines how the body will rotate. We present a physics-informed neural network model to estimate and predict 3D rotational dynamics from image sequences. We achieve this using a multi-stage prediction pipeline that maps individual images to a latent representation homeomorphic to $\mathbf{SO}(3)$, computes angular velocities from latent pairs, and predicts future latent states using the Hamiltonian equations of motion. We demonstrate the efficacy of our approach on new rotating rigid-body datasets of sequences of synthetic images of rotating objects, including cubes, prisms and satellites, with unknown uniform and non-uniform mass distributions.
comment: 21 pages, 9 figures
♻ ☆ The Hidden Language of Diffusion Models
Text-to-image diffusion models have demonstrated an unparalleled ability to generate high-quality, diverse images from a textual concept (e.g., "a doctor", "love"). However, the internal process of mapping text to a rich visual representation remains an enigma. In this work, we tackle the challenge of understanding concept representations in text-to-image models by decomposing an input text prompt into a small set of interpretable elements. This is achieved by learning a pseudo-token that is a sparse weighted combination of tokens from the model's vocabulary, with the objective of reconstructing the images generated for the given concept. Applied over the state-of-the-art Stable Diffusion model, this decomposition reveals non-trivial and surprising structures in the representations of concepts. For example, we find that some concepts such as "a president" or "a composer" are dominated by specific instances (e.g., "Obama", "Biden") and their interpolations. Other concepts, such as "happiness" combine associated terms that can be concrete ("family", "laughter") or abstract ("friendship", "emotion"). In addition to peering into the inner workings of Stable Diffusion, our method also enables applications such as single-image decomposition to tokens, bias detection and mitigation, and semantic image manipulation. Our code will be available at: https://hila-chefer.github.io/Conceptor/
♻ ☆ Scene as Occupancy
Human driver can easily describe the complex traffic scene by visual system. Such an ability of precise perception is essential for driver's planning. To achieve this, a geometry-aware representation that quantizes the physical 3D scene into structured grid map with semantic labels per cell, termed as 3D Occupancy, would be desirable. Compared to the form of bounding box, a key insight behind occupancy is that it could capture the fine-grained details of critical obstacles in the scene, and thereby facilitate subsequent tasks. Prior or concurrent literature mainly concentrate on a single scene completion task, where we might argue that the potential of this occupancy representation might obsess broader impact. In this paper, we propose OccNet, a multi-view vision-centric pipeline with a cascade and temporal voxel decoder to reconstruct 3D occupancy. At the core of OccNet is a general occupancy embedding to represent 3D physical world. Such a descriptor could be applied towards a wide span of driving tasks, including detection, segmentation and planning. To validate the effectiveness of this new representation and our proposed algorithm, we propose OpenOcc, the first dense high-quality 3D occupancy benchmark built on top of nuScenes. Empirical experiments show that there are evident performance gain across multiple tasks, e.g., motion planning could witness a collision rate reduction by 15%-58%, demonstrating the superiority of our method.
♻ ☆ Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding
We present Video-LLaMA, a multi-modal framework that empowers Large Language Models (LLMs) with the capability of understanding both visual and auditory content in the video. Video-LLaMA bootstraps cross-modal training from the frozen pre-trained visual & audio encoders and the frozen LLMs. Unlike previous vision-LLMs that focus on static image comprehensions such as MiniGPT-4 and LLaVA, Video-LLaMA mainly tackles two challenges in video understanding: (1) capturing the temporal changes in visual scenes, (2) integrating audio-visual signals. To counter the first challenge, we propose a Video Q-former to assemble the pre-trained image encoder into our video encoder and introduce a video-to-text generation task to learn video-language correspondence. For the second challenge, we leverage ImageBind, a universal embedding model aligning multiple modalities as the pre-trained audio encoder, and introduce an Audio Q-former on top of ImageBind to learn reasonable auditory query embeddings for the LLM module. To align the output of both visual & audio encoders with LLM's embedding space, we train Video-LLaMA on massive video/image-caption pairs as well as visual-instruction-tuning datasets of moderate amount but higher quality. We found Video-LLaMA showcases the ability to perceive and comprehend video content, generating meaningful responses that are grounded in the visual and auditory information presented in the videos. This highlights the potential of Video-LLaMA as a promising prototype for audio-visual AI assistants.
comment: Technical Report; Code, Pretrained Model, and Dataset: https://github.com/DAMO-NLP-SG/Video-LLaMA
♻ ☆ A bioinspired three-stage model for camouflaged object detection
Camouflaged objects are typically assimilated into their backgrounds and exhibit fuzzy boundaries. The complex environmental conditions and the high intrinsic similarity between camouflaged targets and their surroundings pose significant challenges in accurately locating and segmenting these objects in their entirety. While existing methods have demonstrated remarkable performance in various real-world scenarios, they still face limitations when confronted with difficult cases, such as small targets, thin structures, and indistinct boundaries. Drawing inspiration from human visual perception when observing images containing camouflaged objects, we propose a three-stage model that enables coarse-to-fine segmentation in a single iteration. Specifically, our model employs three decoders to sequentially process subsampled features, cropped features, and high-resolution original features. This proposed approach not only reduces computational overhead but also mitigates interference caused by background noise. Furthermore, considering the significance of multi-scale information, we have designed a multi-scale feature enhancement module that enlarges the receptive field while preserving detailed structural cues. Additionally, a boundary enhancement module has been developed to enhance performance by leveraging boundary information. Subsequently, a mask-guided fusion module is proposed to generate fine-grained results by integrating coarse prediction maps with high-resolution feature maps. Our network surpasses state-of-the-art CNN-based counterparts without unnecessary complexities. Upon acceptance of the paper, the source code will be made publicly available at https://github.com/clelouch/BTSNet.
♻ ☆ Diffusion Models and Semi-Supervised Learners Benefit Mutually with Few Labels
In an effort to further advance semi-supervised generative and classification tasks, we propose a simple yet effective training strategy called dual pseudo training (DPT), built upon strong semi-supervised learners and diffusion models. DPT operates in three stages: training a classifier on partially labeled data to predict pseudo-labels; training a conditional generative model using these pseudo-labels to generate pseudo images; and retraining the classifier with a mix of real and pseudo images. Empirically, DPT consistently achieves SOTA performance of semi-supervised generation and classification across various settings. In particular, with one or two labels per class, DPT achieves a Fr\'echet Inception Distance (FID) score of 3.08 or 2.52 on ImageNet 256x256, surpassing strong diffusion models with full labels, such as IDDPM, CDM, ADM, and LDM. Besides, DPT outperforms competitive semi-supervised baselines substantially on ImageNet classification tasks, achieving top-1 accuracies of 59.0 (+2.8), 69.5 (+3.0), and 74.4 (+2.0) with one, two, or five labels per class, respectively. Notably, our results demonstrate that diffusion can generate realistic images with only a few labels (e.g., <0.1%) and generative augmentation remains viable for semi-supervised classification.
♻ ☆ SELTO: Sample-Efficient Learned Topology Optimization
Recent developments in Deep Learning (DL) suggest a vast potential for Topology Optimization (TO). However, while there are some promising attempts, the subfield still lacks a firm footing regarding basic methods and datasets. We aim to address both points. First, we explore physics-based preprocessing and equivariant networks to create sample-efficient components for TO DL pipelines. We evaluate them in a large-scale ablation study using end-to-end supervised training. The results demonstrate a drastic improvement in sample efficiency and the predictions' physical correctness. Second, to improve comparability and future progress, we publish the two first TO datasets containing problems and corresponding ground truth solutions.
comment: 25 pages, 10 figures, submitted to the International Journal for Numerical Methods in Engineering
♻ ☆ Efficient Multi-Scale Attention Module with Cross-Spatial Learning ICASSP2023
Remarkable effectiveness of the channel or spatial attention mechanisms for producing more discernible feature representation are illustrated in various computer vision tasks. However, modeling the cross-channel relationships with channel dimensionality reduction may bring side effect in extracting deep visual representations. In this paper, a novel efficient multi-scale attention (EMA) module is proposed. Focusing on retaining the information on per channel and decreasing the computational overhead, we reshape the partly channels into the batch dimensions and group the channel dimensions into multiple sub-features which make the spatial semantic features well-distributed inside each feature group. Specifically, apart from encoding the global information to re-calibrate the channel-wise weight in each parallel branch, the output features of the two parallel branches are further aggregated by a cross-dimension interaction for capturing pixel-level pairwise relationship. We conduct extensive ablation studies and experiments on image classification and object detection tasks with popular benchmarks (e.g., CIFAR-100, ImageNet-1k, MS COCO and VisDrone2019) for evaluating its performance.
comment: Accepted to ICASSP2023
♻ ☆ PlaNeRF: SVD Unsupervised 3D Plane Regularization for NeRF Large-Scale Scene Reconstruction
Neural Radiance Fields (NeRF) enable 3D scene reconstruction from 2D images and camera poses for Novel View Synthesis (NVS). Although NeRF can produce photorealistic results, it often suffers from overfitting to training views, leading to poor geometry reconstruction, especially in low-texture areas. This limitation restricts many important applications which require accurate geometry, such as extrapolated NVS, HD mapping and scene editing. To address this limitation, we propose a new method to improve NeRF's 3D structure using only RGB images and semantic maps. Our approach introduces a novel plane regularization based on Singular Value Decomposition (SVD), that does not rely on any geometric prior. In addition, we leverage the Structural Similarity Index Measure (SSIM) in our loss design to properly initialize the volumetric representation of NeRF. Quantitative and qualitative results show that our method outperforms popular regularization approaches in accurate geometry reconstruction for large-scale outdoor scenes and achieves SoTA rendering quality on the KITTI-360 NVS benchmark.
comment: 14 pages, 7 figures
♻ ☆ DiffuseExpand: Expanding dataset for 2D medical image segmentation using diffusion models IJCAI
Dataset expansion can effectively alleviate the problem of data scarcity for medical image segmentation, due to privacy concerns and labeling difficulties. However, existing expansion algorithms still face great challenges due to their inability of guaranteeing the diversity of synthesized images with paired segmentation masks. In recent years, Diffusion Probabilistic Models (DPMs) have shown powerful image synthesis performance, even better than Generative Adversarial Networks. Based on this insight, we propose an approach called DiffuseExpand for expanding datasets for 2D medical image segmentation using DPM, which first samples a variety of masks from Gaussian noise to ensure the diversity, and then synthesizes images to ensure the alignment of images and masks. After that, DiffuseExpand chooses high-quality samples to further enhance the effectiveness of data expansion. Our comparison and ablation experiments on COVID-19 and CGMH Pelvis datasets demonstrate the effectiveness of DiffuseExpand. Our code is released at https://github.com/shaoshitong/DiffuseExpand.
comment: Accepted by IJCAI workshop (1st International Workshop on Generalizing from Limited Resources in the Open World). pre-version was rejected by MICCAI
♻ ☆ RHINO: Rotated DETR with Dynamic Denoising via Hungarian Matching for Oriented Object Detection
With the publication of DINO, a variant of the Detection Transformer (DETR), Detection Transformers are breaking the record in the object detection benchmark with the merits of their end-to-end design and scalability. However, the extension of DETR to oriented object detection has not been thoroughly studied although more benefits from its end-to-end architecture are expected such as removing NMS and anchor-related costs. In this paper, we propose a first strong DINO-based baseline for oriented object detection. We found that straightforward employment of DETRs for oriented object detection does not guarantee non-duplicate prediction, and propose a simple cost to mitigate this. Furthermore, we introduce a $\textit{dynamic denoising}$ strategy that uses Hungarian matching to filter redundant noised queries and $\textit{query alignment}$ to preserve matching consistency between Transformer decoder layers. Our proposed model outperforms previous rotated DETRs and other counterparts, achieving state-of-the-art performance in DOTA-v1.0/v1.5/v2.0, and DIOR-R benchmarks.
comment: State-of-the-art rotated object detector in DOTA v1.0/v1.5/v2.0 and DIOR-R at the time of publication
♻ ☆ Inflated 3D Convolution-Transformer for Weakly-supervised Carotid Stenosis Grading with Ultrasound Videos MICCAI 2023
Localization of the narrowest position of the vessel and corresponding vessel and remnant vessel delineation in carotid ultrasound (US) are essential for carotid stenosis grading (CSG) in clinical practice. However, the pipeline is time-consuming and tough due to the ambiguous boundaries of plaque and temporal variation. To automatize this procedure, a large number of manual delineations are usually required, which is not only laborious but also not reliable given the annotation difficulty. In this study, we present the first video classification framework for automatic CSG. Our contribution is three-fold. First, to avoid the requirement of laborious and unreliable annotation, we propose a novel and effective video classification network for weakly-supervised CSG. Second, to ease the model training, we adopt an inflation strategy for the network, where pre-trained 2D convolution weights can be adapted into the 3D counterpart in our network. In this way, the existing pre-trained large model can be used as an effective warm start for our network. Third, to enhance the feature discrimination of the video, we propose a novel attention-guided multi-dimension fusion (AMDF) transformer encoder to model and integrate global dependencies within and across spatial and temporal dimensions, where two lightweight cross-dimensional attention mechanisms are designed. Our approach is extensively validated on a large clinically collected carotid US video dataset, demonstrating state-of-the-art performance compared with strong competitors.
comment: Accepted by MICCAI 2023
♻ ☆ TEyeD: Over 20 million real-world eye images with Pupil, Eyelid, and Iris 2D and 3D Segmentations, 2D and 3D Landmarks, 3D Eyeball, Gaze Vector, and Eye Movement Types
We present TEyeD, the world's largest unified public data set of eye images taken with head-mounted devices. TEyeD was acquired with seven different head-mounted eye trackers. Among them, two eye trackers were integrated into virtual reality (VR) or augmented reality (AR) devices. The images in TEyeD were obtained from various tasks, including car rides, simulator rides, outdoor sports activities, and daily indoor activities. The data set includes 2D and 3D landmarks, semantic segmentation, 3D eyeball annotation and the gaze vector and eye movement types for all images. Landmarks and semantic segmentation are provided for the pupil, iris and eyelids. Video lengths vary from a few minutes to several hours. With more than 20 million carefully annotated images, TEyeD provides a unique, coherent resource and a valuable foundation for advancing research in the field of computer vision, eye tracking and gaze estimation in modern VR and AR applications. Download: Just connect via FTP as user TEyeDUser and without password to nephrit.cs.uni-tuebingen.de (ftp://nephrit.cs.uni-tuebingen.de).
comment: Download: Just connect via FTP as user TEyeDUser and without password to nephrit.cs.uni-tuebingen.de (ftp://nephrit.cs.uni-tuebingen.de)
♻ ☆ Global Context Vision Transformers ICML 2023
We propose global context vision transformer (GC ViT), a novel architecture that enhances parameter and compute utilization for computer vision. Our method leverages global context self-attention modules, joint with standard local self-attention, to effectively and efficiently model both long and short-range spatial interactions, without the need for expensive operations such as computing attention masks or shifting local windows. In addition, we address the lack of the inductive bias in ViTs, and propose to leverage a modified fused inverted residual blocks in our architecture. Our proposed GC ViT achieves state-of-the-art results across image classification, object detection and semantic segmentation tasks. On ImageNet-1K dataset for classification, the variants of GC ViT with 51M, 90M and 201M parameters achieve 84.3%, 85.0% and 85.7% Top-1 accuracy, respectively, at 224 image resolution and without any pre-training, hence surpassing comparably-sized prior art such as CNN-based ConvNeXt and ViT-based MaxViT and Swin Transformer by a large margin. Pre-trained GC ViT backbones in downstream tasks of object detection, instance segmentation, and semantic segmentation using MS COCO and ADE20K datasets outperform prior work consistently. Specifically, GC ViT with a 4-scale DINO detection head achieves a box AP of 58.3 on MS COCO dataset.
comment: Accepted to ICML 2023
♻ ☆ Benchmarking Spatial Relationships in Text-to-Image Generation
Spatial understanding is a fundamental aspect of computer vision and integral for human-level reasoning about images, making it an important component for grounded language understanding. While recent text-to-image synthesis (T2I) models have shown unprecedented improvements in photorealism, it is unclear whether they have reliable spatial understanding capabilities. We investigate the ability of T2I models to generate correct spatial relationships among objects and present VISOR, an evaluation metric that captures how accurately the spatial relationship described in text is generated in the image. To benchmark existing models, we introduce a dataset, SR2D, that contains sentences describing two objects and the spatial relationship between them. We construct an automated evaluation pipeline to recognize objects and their spatial relationships, and employ it in a large-scale evaluation of T2I models. Our experiments reveal a surprising finding that, although state-of-the-art T2I models exhibit high image quality, they are severely limited in their ability to generate multiple objects or the specified spatial relations between them. Our analyses demonstrate several biases and artifacts of T2I models such as the difficulty with generating multiple objects, a bias towards generating the first object mentioned, spatially inconsistent outputs for equivalent relationships, and a correlation between object co-occurrence and spatial understanding capabilities. We conduct a human study that shows the alignment between VISOR and human judgement about spatial understanding. We offer the SR2D dataset and the VISOR metric to the community in support of T2I reasoning research.
comment: preprint; Code and Data at https://github.com/microsoft/VISOR and https://huggingface.co/datasets/tgokhale/sr2d_visor
♻ ☆ FishRecGAN: An End to End GAN Based Network for Fisheye Rectification and Calibration
We propose an end-to-end deep learning approach to rectify fisheye images and simultaneously calibrate camera intrinsic and distortion parameters. Our method consists of two parts: a Quick Image Rectification Module developed with a Pix2Pix GAN and Wasserstein GAN (W-Pix2PixGAN), and a Calibration Module with a CNN architecture. Our Quick Rectification Network performs robust rectification with good resolution, making it suitable for constant calibration in camera-based surveillance equipment. To achieve high-quality calibration, we use the straightened output from the Quick Rectification Module as a guidance-like semantic feature map for the Calibration Module to learn the geometric relationship between the straightened feature and the distorted feature. We train and validate our method with a large synthesized dataset labeled with well-simulated parameters applied to a perspective image dataset. Our solution has achieved robust performance in high-resolution with a significant PSNR value of 22.343.
♻ ☆ Data Efficient Training with Imbalanced Label Sample Distribution for Fashion Detection
Multi-label classification models have a wide range of applications in E-commerce, including visual-based label predictions and language-based sentiment classifications. A major challenge in achieving satisfactory performance for these tasks in the real world is the notable imbalance in data distribution. For instance, in fashion attribute detection, there may be only six 'puff sleeve' clothes among 1000 products in most E-commerce fashion catalogs. To address this issue, we explore more data-efficient model training techniques rather than acquiring a huge amount of annotations to collect sufficient samples, which is neither economic nor scalable. In this paper, we propose a state-of-the-art weighted objective function to boost the performance of deep neural networks (DNNs) for multi-label classification with long-tailed data distribution. Our experiments involve image-based attribute classification of fashion apparels, and the results demonstrate favorable performance for the new weighting method compared to non-weighted and inverse-frequency-based weighting mechanisms. We further evaluate the robustness of the new weighting mechanism using two popular fashion attribute types in today's fashion industry: sleevetype and archetype.
comment: We have identified a substantial error in the experimental results and a potentially misleading explanation of the algorithm. We kindly request that you consider withdrawing this version to mitigate the risk of disseminating inaccurate information
♻ ☆ Cold PAWS: Unsupervised class discovery and addressing the cold-start problem for semi-supervised learning
In many machine learning applications, labeling datasets can be an arduous and time-consuming task. Although research has shown that semi-supervised learning techniques can achieve high accuracy with very few labels within the field of computer vision, little attention has been given to how images within a dataset should be selected for labeling. In this paper, we propose a novel approach based on well-established self-supervised learning, clustering, and manifold learning techniques that address this challenge of selecting an informative image subset to label in the first instance, which is known as the cold-start or unsupervised selective labelling problem. We test our approach using several publicly available datasets, namely CIFAR10, Imagenette, DeepWeeds, and EuroSAT, and observe improved performance with both supervised and semi-supervised learning strategies when our label selection strategy is used, in comparison to random sampling. We also obtain superior performance for the datasets considered with a much simpler approach compared to other methods in the literature.
♻ ☆ Deep learning for diffusion in porous media
We adopt convolutional neural networks (CNN) to predict the basic properties of the porous media. Two different media types are considered: one mimics the sand packings, and the other mimics the systems derived from the extracellular space of biological tissues. The Lattice Boltzmann Method is used to obtain the labeled data necessary for performing supervised learning. We distinguish two tasks. In the first, networks based on the analysis of the system's geometry predict porosity and effective diffusion coefficient. In the second, networks reconstruct the concentration map. In the first task, we propose two types of CNN models: the C-Net and the encoder part of the U-Net. Both networks are modified by adding a self-normalization module [Graczyk \textit{et al.}, Sci Rep 12, 10583 (2022)]. The models predict with reasonable accuracy but only within the data type, they are trained on. For instance, the model trained on sand packings-like samples overshoots or undershoots for biological-like samples. In the second task, we propose the usage of the U-Net architecture. It accurately reconstructs the concentration fields. In contrast to the first task, the network trained on one data type works well for the other. For instance, the model trained on sand packings-like samples works perfectly on biological-like samples. Eventually, for both types of the data, we fit exponents in the Archie's law to find tortuosity that is used to describe the dependence of the effective diffusion on porosity.
comment: 15 pages, 17 figures, to appear in Sci. Rep
♻ ☆ Entropy-driven Unsupervised Keypoint Representation Learning in Videos ICML 2023
Extracting informative representations from videos is fundamental for effectively learning various downstream tasks. We present a novel approach for unsupervised learning of meaningful representations from videos, leveraging the concept of image spatial entropy (ISE) that quantifies the per-pixel information in an image. We argue that \textit{local entropy} of pixel neighborhoods and their temporal evolution create valuable intrinsic supervisory signals for learning prominent features. Building on this idea, we abstract visual features into a concise representation of keypoints that act as dynamic information transmitters, and design a deep learning model that learns, purely unsupervised, spatially and temporally consistent representations \textit{directly} from video frames. Two original information-theoretic losses, computed from local entropy, guide our model to discover consistent keypoint representations; a loss that maximizes the spatial information covered by the keypoints and a loss that optimizes the keypoints' information transportation over time. We compare our keypoint representation to strong baselines for various downstream tasks, \eg, learning object dynamics. Our empirical results show superior performance for our information-driven keypoints that resolve challenges like attendance to static and dynamic objects or objects abruptly entering and leaving the scene.
comment: 29 pages, 14 figures, Accepted at ICML 2023
♻ ☆ FishEye8K: A Benchmark and Dataset for Fisheye Camera Object Detection CVPR
With the advance of AI, road object detection has been a prominent topic in computer vision, mostly using perspective cameras. Fisheye lens provides omnidirectional wide coverage for using fewer cameras to monitor road intersections, however with view distortions. To our knowledge, there is no existing open dataset prepared for traffic surveillance on fisheye cameras. This paper introduces an open FishEye8K benchmark dataset for road object detection tasks, which comprises 157K bounding boxes across five classes (Pedestrian, Bike, Car, Bus, and Truck). In addition, we present benchmark results of State-of-The-Art (SoTA) models, including variations of YOLOv5, YOLOR, YOLO7, and YOLOv8. The dataset comprises 8,000 images recorded in 22 videos using 18 fisheye cameras for traffic monitoring in Hsinchu, Taiwan, at resolutions of 1080$\times$1080 and 1280$\times$1280. The data annotation and validation process were arduous and time-consuming, due to the ultra-wide panoramic and hemispherical fisheye camera images with large distortion and numerous road participants, particularly people riding scooters. To avoid bias, frames from a particular camera were assigned to either the training or test sets, maintaining a ratio of about 70:30 for both the number of images and bounding boxes in each class. Experimental results show that YOLOv8 and YOLOR outperform on input sizes 640$\times$640 and 1280$\times$1280, respectively. The dataset will be available on GitHub with PASCAL VOC, MS COCO, and YOLO annotation formats. The FishEye8K benchmark will provide significant contributions to the fisheye video analytics and smart city applications.
comment: CVPR Workshops 2023
♻ ☆ When are Post-hoc Conceptual Explanations Identifiable? UAI2023
Interest in understanding and factorizing learned embedding spaces through conceptual explanations is steadily growing. When no human concept labels are available, concept discovery methods search trained embedding spaces for interpretable concepts like object shape or color that can provide post-hoc explanations for decisions. Unlike previous work, we argue that concept discovery should be identifiable, meaning that a number of known concepts can be provably recovered to guarantee reliability of the explanations. As a starting point, we explicitly make the connection between concept discovery and classical methods like Principal Component Analysis and Independent Component Analysis by showing that they can recover independent concepts under non-Gaussian distributions. For dependent concepts, we propose two novel approaches that exploit functional compositionality properties of image-generating processes. Our provably identifiable concept discovery methods substantially outperform competitors on a battery of experiments including hundreds of trained models and dependent concepts, where they exhibit up to 29 % better alignment with the ground truth. Our results highlight the strict conditions under which reliable concept discovery without human labels can be guaranteed and provide a formal foundation for the domain. Our code is available online.
comment: v5: UAI2023 camera-ready including supplementary material. The first two authors contributed equally
♻ ☆ Style Interleaved Learning for Generalizable Person Re-identification
Domain generalization (DG) for person re-identification (ReID) is a challenging problem, as access to target domain data is not permitted during the training process. Most existing DG ReID methods update the feature extractor and classifier parameters based on the same features. This common practice causes the model to overfit to existing feature styles in the source domain, resulting in sub-optimal generalization ability on target domains. To solve this problem, we propose a novel style interleaved learning (IL) framework. Unlike conventional learning strategies, IL incorporates two forward propagations and one backward propagation for each iteration. We employ the features of interleaved styles to update the feature extractor and classifiers using different forward propagations, which helps to prevent the model from overfitting to certain domain styles. To generate interleaved feature styles, we further propose a new feature stylization approach. It produces a wide range of meaningful styles that are both different and independent from the original styles in the source domain, which caters to the IL methodology. Extensive experimental results show that our model not only consistently outperforms state-of-the-art methods on large-scale benchmarks for DG ReID, but also has clear advantages in computational efficiency. The code is available at https://github.com/WentaoTan/Interleaved-Learning.
♻ ☆ Towards Unified Text-based Person Retrieval: A Large-scale Multi-Attribute and Language Search Benchmark
In this paper, we introduce a large Multi-Attribute and Language Search dataset for text-based person retrieval, called MALS, and explore the feasibility of performing pre-training on both attribute recognition and image-text matching tasks in one stone. In particular, MALS contains 1,510,330 image-text pairs, which is about 37.5 times larger than prevailing CUHK-PEDES, and all images are annotated with 27 attributes. Considering the privacy concerns and annotation costs, we leverage the off-the-shelf diffusion models to generate the dataset. To verify the feasibility of learning from the generated data, we develop a new joint Attribute Prompt Learning and Text Matching Learning (APTM) framework, considering the shared knowledge between attribute and text. As the name implies, APTM contains an attribute prompt learning stream and a text matching learning stream. (1) The attribute prompt learning leverages the attribute prompts for image-attribute alignment, which enhances the text matching learning. (2) The text matching learning facilitates the representation learning on fine-grained details, and in turn, boosts the attribute prompt learning. Extensive experiments validate the effectiveness of the pre-training on MALS, achieving state-of-the-art retrieval performance via APTM on three challenging real-world benchmarks. In particular, APTM achieves a consistent improvement of +6.60%, +7.39%, and +15.90% Recall@1 accuracy on CUHK-PEDES, ICFG-PEDES, and RSTPReid datasets by a clear margin, respectively.
♻ ☆ DuDGAN: Improving Class-Conditional GANs via Dual-Diffusion
Class-conditional image generation using generative adversarial networks (GANs) has been investigated through various techniques; however, it continues to face challenges such as mode collapse, training instability, and low-quality output in cases of datasets with high intra-class variation. Furthermore, most GANs often converge in larger iterations, resulting in poor iteration efficacy in training procedures. While Diffusion-GAN has shown potential in generating realistic samples, it has a critical limitation in generating class-conditional samples. To overcome these limitations, we propose a novel approach for class-conditional image generation using GANs called DuDGAN, which incorporates a dual diffusion-based noise injection process. Our method consists of three unique networks: a discriminator, a generator, and a classifier. During the training process, Gaussian-mixture noises are injected into the two noise-aware networks, the discriminator and the classifier, in distinct ways. This noisy data helps to prevent overfitting by gradually introducing more challenging tasks, leading to improved model performance. As a result, our method outperforms state-of-the-art conditional GAN models for image generation in terms of performance. We evaluated our method using the AFHQ, Food-101, and CIFAR-10 datasets and observed superior results across metrics such as FID, KID, Precision, and Recall score compared with comparison models, highlighting the effectiveness of our approach.
comment: 8 page, 3 figures, supplementary material included
♻ ☆ Adversarial Example Does Good: Preventing Painting Imitation from Diffusion Models via Adversarial Examples ICML2023
Recently, Diffusion Models (DMs) boost a wave in AI for Art yet raise new copyright concerns, where infringers benefit from using unauthorized paintings to train DMs to generate novel paintings in a similar style. To address these emerging copyright violations, in this paper, we are the first to explore and propose to utilize adversarial examples for DMs to protect human-created artworks. Specifically, we first build a theoretical framework to define and evaluate the adversarial examples for DMs. Then, based on this framework, we design a novel algorithm, named AdvDM, which exploits a Monte-Carlo estimation of adversarial examples for DMs by optimizing upon different latent variables sampled from the reverse process of DMs. Extensive experiments show that the generated adversarial examples can effectively hinder DMs from extracting their features. Therefore, our method can be a powerful tool for human artists to protect their copyright against infringers equipped with DM-based AI-for-Art applications. The code of our method is available on GitHub: https://github.com/mist-project/mist.git.
comment: Accepted by ICML2023 (Oral)
♻ ☆ Human-in-the-loop Embodied Intelligence with Interactive Simulation Environment for Surgical Robot Learning
Surgical robot automation has attracted increasing research interest over the past decade, expecting its potential to benefit surgeons, nurses and patients. Recently, the learning paradigm of embodied intelligence has demonstrated promising ability to learn good control policies for various complex tasks, where embodied AI simulators play an essential role to facilitate relevant research. However, existing open-sourced simulators for surgical robot are still not sufficiently supporting human interactions through physical input devices, which further limits effective investigations on how the human demonstrations would affect policy learning. In this work, we study human-in-the-loop embodied intelligence with a new interactive simulation platform for surgical robot learning. Specifically, we establish our platform based on our previously released SurRoL simulator with several new features co-developed to allow high-quality human interaction via an input device. We showcase the improvement of our simulation environment with the designed new features, and validate effectiveness of incorporating human factors in embodied intelligence through the use of human demonstrations and reinforcement learning as a representative example. Promising results are obtained in terms of learning efficiency. Lastly, five new surgical robot training tasks are developed and released, with which we hope to pave the way for future research on surgical embodied intelligence. Our learning platform is publicly released and will be continuously updated in the website: https://med-air.github.io/SurRoL.
♻ ☆ Interaction Visual Transformer for Egocentric Action Anticipation
Human-object interaction is one of the most important visual cues and we propose a novel way to represent human-object interactions for egocentric action anticipation. We propose a novel transformer variant to model interactions by computing the change in the appearance of objects and human hands due to the execution of the actions and use those changes to refine the video representation. Specifically, we model interactions between hands and objects using Spatial Cross-Attention (SCA) and further infuse contextual information using Trajectory Cross-Attention to obtain environment-refined interaction tokens. Using these tokens, we construct an interaction-centric video representation for action anticipation. We term our model InAViT which achieves state-of-the-art action anticipation performance on large-scale egocentric datasets EPICKTICHENS100 (EK100) and EGTEA Gaze+. InAViT outperforms other visual transformer-based methods including object-centric video representation. On the EK100 evaluation server, InAViT is the top-performing method on the public leaderboard (at the time of submission) where it outperforms the second-best model by 3.3% on mean-top5 recall.
comment: fixed affliation
♻ ☆ Predicting the Next Action by Modeling the Abstract Goal
The problem of anticipating human actions is an inherently uncertain one. However, we can reduce this uncertainty if we have a sense of the goal that the actor is trying to achieve. Here, we present an action anticipation model that leverages goal information for the purpose of reducing the uncertainty in future predictions. Since we do not possess goal information or the observed actions during inference, we resort to visual representation to encapsulate information about both actions and goals. Through this, we derive a novel concept called abstract goal which is conditioned on observed sequences of visual features for action anticipation. We design the abstract goal as a distribution whose parameters are estimated using a variational recurrent network. We sample multiple candidates for the next action and introduce a goal consistency measure to determine the best candidate that follows from the abstract goal. Our method obtains impressive results on the very challenging Epic-Kitchens55 (EK55), EK100, and EGTEA Gaze+ datasets. We obtain absolute improvements of +13.69, +11.24, and +5.19 for Top-1 verb, Top-1 noun, and Top-1 action anticipation accuracy respectively over prior state-of-the-art methods for seen kitchens (S1) of EK55. Similarly, we also obtain significant improvements in the unseen kitchens (S2) set for Top-1 verb (+10.75), noun (+5.84) and action (+2.87) anticipation. Similar trend is observed for EGTEA Gaze+ dataset, where absolute improvement of +9.9, +13.1 and +6.8 is obtained for noun, verb, and action anticipation. It is through the submission of this paper that our method is currently the new state-of-the-art for action anticipation in EK55 and EGTEA Gaze+ https://competitions.codalab.org/competitions/20071#results Code available at https://github.com/debadityaroy/Abstract_Goal
comment: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible
♻ ☆ Patch-Mix Transformer for Unsupervised Domain Adaptation: A Game Perspective CVPR 2023
Endeavors have been recently made to leverage the vision transformer (ViT) for the challenging unsupervised domain adaptation (UDA) task. They typically adopt the cross-attention in ViT for direct domain alignment. However, as the performance of cross-attention highly relies on the quality of pseudo labels for targeted samples, it becomes less effective when the domain gap becomes large. We solve this problem from a game theory's perspective with the proposed model dubbed as PMTrans, which bridges source and target domains with an intermediate domain. Specifically, we propose a novel ViT-based module called PatchMix that effectively builds up the intermediate domain, i.e., probability distribution, by learning to sample patches from both domains based on the game-theoretical models. This way, it learns to mix the patches from the source and target domains to maximize the cross entropy (CE), while exploiting two semi-supervised mixup losses in the feature and label spaces to minimize it. As such, we interpret the process of UDA as a min-max CE game with three players, including the feature extractor, classifier, and PatchMix, to find the Nash Equilibria. Moreover, we leverage attention maps from ViT to re-weight the label of each patch by its importance, making it possible to obtain more domain-discriminative feature representations. We conduct extensive experiments on four benchmark datasets, and the results show that PMTrans significantly surpasses the ViT-based and CNN-based SoTA methods by +3.6% on Office-Home, +1.4% on Office-31, and +17.7% on DomainNet, respectively.
comment: Accepted by CVPR 2023 (Highlight)
♻ ☆ Towards Better Explanations for Object Detection
Recent advances in Artificial Intelligence (AI) technology have promoted their use in almost every field. The growing complexity of deep neural networks (DNNs) makes it increasingly difficult and important to explain the inner workings and decisions of the network. However, most current techniques for explaining DNNs focus mainly on interpreting classification tasks. This paper proposes a method to explain the decision for any object detection model called D-CLOSE. To closely track the model's behavior, we used multiple levels of segmentation on the image and a process to combine them. We performed tests on the MS-COCO dataset with the YOLOX model, which shows that our method outperforms D-RISE and can give a better quality and less noise explanation.
comment: 9 pages, 10 figures
♻ ☆ Stable Diffusion is Unstable
Recently, text-to-image models have been thriving. Despite their powerful generative capacity, our research has uncovered a lack of robustness in this generation process. Specifically, the introduction of small perturbations to the text prompts can result in the blending of primary subjects with other categories or their complete disappearance in the generated images. In this paper, we propose Auto-attack on Text-to-image Models (ATM), a gradient-based approach, to effectively and efficiently generate such perturbations. By learning a Gumbel Softmax distribution, we can make the discrete process of word replacement or extension continuous, thus ensuring the differentiability of the perturbation generation. Once the distribution is learned, ATM can sample multiple attack samples simultaneously. These attack samples can prevent the generative model from generating the desired subjects without compromising image quality. ATM has achieved a 91.1% success rate in short-text attacks and an 81.2% success rate in long-text attacks. Further empirical analysis revealed four attack patterns based on: 1) the variability in generation speed, 2) the similarity of coarse-grained characteristics, 3) the polysemy of words, and 4) the positioning of words.
comment: 22 pages, 20 figures
♻ ☆ Long-range UAV Thermal Geo-localization with Satellite Imagery
Onboard sensors, such as cameras and thermal sensors, have emerged as effective alternatives to Global Positioning System (GPS) for geo-localization in Unmanned Aerial Vehicle (UAV) navigation. Since GPS can suffer from signal loss and spoofing problems, researchers have explored camera-based techniques such as Visual Geo-localization (VG) using satellite imagery. Additionally, thermal geo-localization (TG) has become crucial for long-range UAV flights in low-illumination environments. This paper proposes a novel thermal geo-localization framework using satellite imagery, which includes multiple domain adaptation methods to address the limited availability of paired thermal and satellite images. The experimental results demonstrate the effectiveness of the proposed approach in achieving reliable thermal geo-localization performance, even in thermal images with indistinct self-similar features. We evaluate our approach on real data collected onboard a UAV. We also release the code and \textit{Boson-nighttime}, a dataset of paired satellite-thermal and unpaired satellite images for thermal geo-localization with satellite imagery. To the best of our knowledge, this work is the first to propose a thermal geo-localization method using satellite imagery in long-range flights.
comment: 8 pages, 6 figures
♻ ☆ Nonparametric Iterative Machine Teaching ICML 2023
In this paper, we consider the problem of Iterative Machine Teaching (IMT), where the teacher provides examples to the learner iteratively such that the learner can achieve fast convergence to a target model. However, existing IMT algorithms are solely based on parameterized families of target models. They mainly focus on convergence in the parameter space, resulting in difficulty when the target models are defined to be functions without dependency on parameters. To address such a limitation, we study a more general task -- Nonparametric Iterative Machine Teaching (NIMT), which aims to teach nonparametric target models to learners in an iterative fashion. Unlike parametric IMT that merely operates in the parameter space, we cast NIMT as a functional optimization problem in the function space. To solve it, we propose both random and greedy functional teaching algorithms. We obtain the iterative teaching dimension (ITD) of the random teaching algorithm under proper assumptions, which serves as a uniform upper bound of ITD in NIMT. Further, the greedy teaching algorithm has a significantly lower ITD, which reaches a tighter upper bound of ITD in NIMT. Finally, we verify the correctness of our theoretical findings with extensive experiments in nonparametric scenarios.
comment: ICML 2023 (20 pages, 10 figures)
♻ ☆ VideoComposer: Compositional Video Synthesis with Motion Controllability
The pursuit of controllability as a higher standard of visual content creation has yielded remarkable progress in customizable image synthesis. However, achieving controllable video synthesis remains challenging due to the large variation of temporal dynamics and the requirement of cross-frame temporal consistency. Based on the paradigm of compositional generation, this work presents VideoComposer that allows users to flexibly compose a video with textual conditions, spatial conditions, and more importantly temporal conditions. Specifically, considering the characteristic of video data, we introduce the motion vector from compressed videos as an explicit control signal to provide guidance regarding temporal dynamics. In addition, we develop a Spatio-Temporal Condition encoder (STC-encoder) that serves as a unified interface to effectively incorporate the spatial and temporal relations of sequential inputs, with which the model could make better use of temporal conditions and hence achieve higher inter-frame consistency. Extensive experimental results suggest that VideoComposer is able to control the spatial and temporal patterns simultaneously within a synthesized video in various forms, such as text description, sketch sequence, reference video, or even simply hand-crafted motions. The code and models will be publicly available at https://videocomposer.github.io.
comment: The first four authors contributed equally. Project page: https://videocomposer.github.io
♻ ☆ Marching-Primitives: Shape Abstraction from Signed Distance Function CVPR2023
Representing complex objects with basic geometric primitives has long been a topic in computer vision. Primitive-based representations have the merits of compactness and computational efficiency in higher-level tasks such as physics simulation, collision checking, and robotic manipulation. Unlike previous works which extract polygonal meshes from a signed distance function (SDF), in this paper, we present a novel method, named Marching-Primitives, to obtain a primitive-based abstraction directly from an SDF. Our method grows geometric primitives (such as superquadrics) iteratively by analyzing the connectivity of voxels while marching at different levels of signed distance. For each valid connected volume of interest, we march on the scope of voxels from which a primitive is able to be extracted in a probabilistic sense and simultaneously solve for the parameters of the primitive to capture the underlying local geometry. We evaluate the performance of our method on both synthetic and real-world datasets. The results show that the proposed method outperforms the state-of-the-art in terms of accuracy, and is directly generalizable among different categories and scales. The code is open-sourced at https://github.com/ChirikjianLab/Marching-Primitives.git.
comment: Accepted to CVPR2023 Highlight
♻ ☆ Scaling Up 3D Kernels with Bayesian Frequency Re-parameterization for Medical Image Segmentation MICCAI 2023
With the inspiration of vision transformers, the concept of depth-wise convolution revisits to provide a large Effective Receptive Field (ERF) using Large Kernel (LK) sizes for medical image segmentation. However, the segmentation performance might be saturated and even degraded as the kernel sizes scaled up (e.g., $21\times 21\times 21$) in a Convolutional Neural Network (CNN). We hypothesize that convolution with LK sizes is limited to maintain an optimal convergence for locality learning. While Structural Re-parameterization (SR) enhances the local convergence with small kernels in parallel, optimal small kernel branches may hinder the computational efficiency for training. In this work, we propose RepUX-Net, a pure CNN architecture with a simple large kernel block design, which competes favorably with current network state-of-the-art (SOTA) (e.g., 3D UX-Net, SwinUNETR) using 6 challenging public datasets. We derive an equivalency between kernel re-parameterization and the branch-wise variation in kernel convergence. Inspired by the spatial frequency in the human visual system, we extend to vary the kernel convergence into element-wise setting and model the spatial frequency as a Bayesian prior to re-parameterize convolutional weights during training. Specifically, a reciprocal function is leveraged to estimate a frequency-weighted value, which rescales the corresponding kernel element for stochastic gradient descent. From the experimental results, RepUX-Net consistently outperforms 3D SOTA benchmarks with internal validation (FLARE: 0.929 to 0.944), external validation (MSD: 0.901 to 0.932, KiTS: 0.815 to 0.847, LiTS: 0.933 to 0.949, TCIA: 0.736 to 0.779) and transfer learning (AMOS: 0.880 to 0.911) scenarios in Dice Score.
comment: Accepted to MICCAI 2023 (top 13.6%), both codes and pretrained models are available at: https://github.com/MASILab/RepUX-Net
♻ ☆ Comments on 'Fast and scalable search of whole-slide images via self-supervised deep learning'
Chen et al. [Chen2022] recently published the article 'Fast and scalable search of whole-slide images via self-supervised deep learning' in Nature Biomedical Engineering. The authors call their method 'self-supervised image search for histology', short SISH. We express our concerns that SISH is an incremental modification of Yottixel, has used MinMax binarization but does not cite the original works, and is based on a misnomer 'self-supervised image search'. As well, we point to several other concerns regarding experiments and comparisons performed by Chen et al.
♻ ☆ PDFVQA: A New Dataset for Real-World VQA on PDF Documents ECML-PKDD 2023
Document-based Visual Question Answering examines the document understanding of document images in conditions of natural language questions. We proposed a new document-based VQA dataset, PDF-VQA, to comprehensively examine the document understanding from various aspects, including document element recognition, document layout structural understanding as well as contextual understanding and key information extraction. Our PDF-VQA dataset extends the current scale of document understanding that limits on the single document page to the new scale that asks questions over the full document of multiple pages. We also propose a new graph-based VQA model that explicitly integrates the spatial and hierarchically structural relationships between different document elements to boost the document structural understanding. The performances are compared with several baselines over different question types and tasks\footnote{The full dataset will be released after paper acceptance.
comment: Accepted by ECML-PKDD 2023
♻ ☆ Blind Image Deconvolution Using Variational Deep Image Prior
Conventional deconvolution methods utilize hand-crafted image priors to constrain the optimization. While deep-learning-based methods have simplified the optimization by end-to-end training, they fail to generalize well to blurs unseen in the training dataset. Thus, training image-specific models is important for higher generalization. Deep image prior (DIP) provides an approach to optimize the weights of a randomly initialized network with a single degraded image by maximum a posteriori (MAP), which shows that the architecture of a network can serve as the hand-crafted image prior. Different from the conventional hand-crafted image priors that are statistically obtained, it is hard to find a proper network architecture because the relationship between images and their corresponding network architectures is unclear. As a result, the network architecture cannot provide enough constraint for the latent sharp image. This paper proposes a new variational deep image prior (VDIP) for blind image deconvolution, which exploits additive hand-crafted image priors on latent sharp images and approximates a distribution for each pixel to avoid suboptimal solutions. Our mathematical analysis shows that the proposed method can better constrain the optimization. The experimental results further demonstrate that the generated images have better quality than that of the original DIP on benchmark datasets. The source code of our VDIP is available at https://github.com/Dong-Huo/VDIP-Deconvolution.
comment: IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2023
♻ ☆ Training Like a Medical Resident: Universal Medical Image Segmentation via Context Prior Learning
A major enduring focus of clinical workflows is disease analytics and diagnosis, leading to medical imaging datasets where the modalities and annotations are strongly tied to specific clinical objectives. To date, building task-specific segmentation models is intuitive yet a restrictive approach, lacking insights gained from widespread imaging cohorts. Inspired by the training of medical residents, we explore universal medical image segmentation, whose goal is to learn from diverse medical imaging sources covering a range of clinical targets, body regions, and image modalities. Following this paradigm, we propose Hermes, a context prior learning approach that addresses the challenges related to the heterogeneity on data, modality, and annotations in the proposed universal paradigm. In a collection of seven diverse datasets, we demonstrate the appealing merits of the universal paradigm over the traditional task-specific training paradigm. By leveraging the synergy among various tasks, Hermes shows superior performance and model scalability. Our in-depth investigation on two additional datasets reveals Hermes' strong capabilities for transfer learning, incremental learning, and generalization to different downstream tasks. The code is available: https://github.com/yhygao/universal-medical-image-segmentation.
♻ ☆ Learned Alternating Minimization Algorithm for Dual-domain Sparse-View CT Reconstruction
We propose a novel Learned Alternating Minimization Algorithm (LAMA) for dual-domain sparse-view CT image reconstruction. LAMA is naturally induced by a variational model for CT reconstruction with learnable nonsmooth nonconvex regularizers, which are parameterized as composite functions of deep networks in both image and sinogram domains. To minimize the objective of the model, we incorporate the smoothing technique and residual learning architecture into the design of LAMA. We show that LAMA substantially reduces network complexity, improves memory efficiency and reconstruction accuracy, and is provably convergent for reliable reconstructions. Extensive numerical experiments demonstrate that LAMA outperforms existing methods by a wide margin on multiple benchmark CT datasets.
♻ ☆ 3rd Place Solution for PVUW2023 VSS Track: A Large Model for Semantic Segmentation on VSPW CVPR 2023
In this paper, we introduce 3rd place solution for PVUW2023 VSS track. Semantic segmentation is a fundamental task in computer vision with numerous real-world applications. We have explored various image-level visual backbones and segmentation heads to tackle the problem of video semantic segmentation. Through our experimentation, we find that InternImage-H as the backbone and Mask2former as the segmentation head achieves the best performance. In addition, we explore two post-precessing methods: CascadePSP and Segment Anything Model (SAM). Ultimately, our approach obtains 62.60\% and 64.84\% mIoU on the VSPW test set1 and final test set, respectively, securing the third position in the PVUW2023 VSS track.
comment: 3rd Place Solution for CVPR 2023 PVUW VSS Track
♻ ☆ Hyperbolic Image-Text Representations ICML 2023
Visual and linguistic concepts naturally organize themselves in a hierarchy, where a textual concept "dog" entails all images that contain dogs. Despite being intuitive, current large-scale vision and language models such as CLIP do not explicitly capture such hierarchy. We propose MERU, a contrastive model that yields hyperbolic representations of images and text. Hyperbolic spaces have suitable geometric properties to embed tree-like data, so MERU can better capture the underlying hierarchy in image-text datasets. Our results show that MERU learns a highly interpretable and structured representation space while being competitive with CLIP's performance on standard multi-modal tasks like image classification and image-text retrieval.
comment: ICML 2023
♻ ☆ Retrieval-Augmented Multimodal Language Modeling ICML 2023
Recent multimodal models such as DALL-E and CM3 have achieved remarkable progress in text-to-image and image-to-text generation. However, these models store all learned knowledge (e.g., the appearance of the Eiffel Tower) in the model parameters, requiring increasingly larger models and training data to capture more knowledge. To integrate knowledge in a more scalable and modular way, we propose a retrieval-augmented multimodal model, which enables a base multimodal model (generator) to refer to relevant text and images fetched by a retriever from external memory (e.g., documents on the web). Specifically, for the retriever, we use a pretrained CLIP, and for the generator, we train a CM3 Transformer on the LAION dataset. Our resulting model, named Retrieval-Augmented CM3 (RA-CM3), is the first multimodal model that can retrieve and generate both text and images. We show that RA-CM3 significantly outperforms baseline multimodal models such as DALL-E and CM3 on both image and caption generation tasks (12 FID and 17 CIDEr improvements on MS-COCO), while requiring much less compute for training (<30% of DALL-E). Moreover, we show that RA-CM3 exhibits novel capabilities, such as faithful image generation and multimodal in-context learning (e.g., image generation from demonstrations).
comment: Published at ICML 2023. Blog post available at https://cs.stanford.edu/~myasu/blog/racm3/
♻ ☆ Building Floorspace in China: A Dataset and Learning Pipeline
This paper provides a first milestone in measuring the floorspace of buildings (that is, building footprint and height) for 40 major Chinese cities. The intent is to maximize city coverage and, eventually provide longitudinal data. Doing so requires building on imagery that is of a medium-fine-grained granularity, as larger cross sections of cities and longer time series for them are only available in such format. We use a multi-task object segmenter approach to learn the building footprint and height in the same framework in parallel: (1) we determine the surface area is covered by any buildings (the square footage of occupied land); (2) we determine floorspace from multi-image representations of buildings from various angles to determine the height of buildings. We use Sentinel-1 and -2 satellite images as our main data source. The benefits of these data are their large cross-sectional and longitudinal scope plus their unrestricted accessibility. We provide a detailed description of our data, algorithms, and evaluations. In addition, we analyze the quality of reference data and their role for measuring the building floorspace with minimal error. We conduct extensive quantitative and qualitative analyses with Shenzhen as a case study using our multi-task learner. Finally, we conduct correlation studies between our results (on both pixel and aggregated urban area levels) and nightlight data to gauge the merits of our approach in studying urban development. Our data and codebase are publicly accessible under https://gitlab.ethz.ch/raox/urban-satellite-public-v2.
♻ ☆ SinDDM: A Single Image Denoising Diffusion Model ICML 2023
Denoising diffusion models (DDMs) have led to staggering performance leaps in image generation, editing and restoration. However, existing DDMs use very large datasets for training. Here, we introduce a framework for training a DDM on a single image. Our method, which we coin SinDDM, learns the internal statistics of the training image by using a multi-scale diffusion process. To drive the reverse diffusion process, we use a fully-convolutional light-weight denoiser, which is conditioned on both the noise level and the scale. This architecture allows generating samples of arbitrary dimensions, in a coarse-to-fine manner. As we illustrate, SinDDM generates diverse high-quality samples, and is applicable in a wide array of tasks, including style transfer and harmonization. Furthermore, it can be easily guided by external supervision. Particularly, we demonstrate text-guided generation from a single image using a pre-trained CLIP model.
comment: Updated for ICML 2023 and added the Appendix. Note that the images are lightly compressed. Visit our project page for uncompressed results: https://matankleiner.github.io/sinddm/
♻ ☆ DisC-Diff: Disentangled Conditional Diffusion Model for Multi-Contrast MRI Super-Resolution MICCAI 2023
Multi-contrast magnetic resonance imaging (MRI) is the most common management tool used to characterize neurological disorders based on brain tissue contrasts. However, acquiring high-resolution MRI scans is time-consuming and infeasible under specific conditions. Hence, multi-contrast super-resolution methods have been developed to improve the quality of low-resolution contrasts by leveraging complementary information from multi-contrast MRI. Current deep learning-based super-resolution methods have limitations in estimating restoration uncertainty and avoiding mode collapse. Although the diffusion model has emerged as a promising approach for image enhancement, capturing complex interactions between multiple conditions introduced by multi-contrast MRI super-resolution remains a challenge for clinical applications. In this paper, we propose a disentangled conditional diffusion model, DisC-Diff, for multi-contrast brain MRI super-resolution. It utilizes the sampling-based generation and simple objective function of diffusion models to estimate uncertainty in restorations effectively and ensure a stable optimization process. Moreover, DisC-Diff leverages a disentangled multi-stream network to fully exploit complementary information from multi-contrast MRI, improving model interpretation under multiple conditions of multi-contrast inputs. We validated the effectiveness of DisC-Diff on two datasets: the IXI dataset, which contains 578 normal brains, and a clinical dataset with 316 pathological brains. Our experimental results demonstrate that DisC-Diff outperforms other state-of-the-art methods both quantitatively and visually.
comment: Early Accepted by MICCAI 2023
♻ ☆ Statistical shape representations for temporal registration of plant components in 3D ICRA 2023
Plants are dynamic organisms and understanding temporal variations in vegetation is an essential problem for robots in the wild. However, associating repeated 3D scans of plants across time is challenging. A key step in this process is re-identifying and tracking the same individual plant components over time. Previously, this has been achieved by comparing their global spatial or topological location. In this work, we demonstrate how using shape features improves temporal organ matching. We present a landmark-free shape compression algorithm, which allows for the extraction of 3D shape features of leaves, characterises leaf shape and curvature efficiently in few parameters, and makes the association of individual leaves in feature space possible. The approach combines 3D contour extraction and further compression using Principal Component Analysis (PCA) to produce a shape space encoding, which is entirely learned from data and retains information about edge contours and 3D curvature. Our evaluation on temporal scan sequences of tomato plants shows, that incorporating shape features improves temporal leaf-matching. A combination of shape, location, and rotation information proves most informative for recognition of leaves over time and yields a true positive rate of 75%, a 15% improvement on sate-of-the-art methods. This is essential for robotic crop monitoring, which enables whole-of-lifecycle phenotyping.
comment: 6 pages plus references, 7 figures, presented at ICRA 2023
♻ ☆ Rethinking Robust Contrastive Learning from the Adversarial Perspective
To advance the understanding of robust deep learning, we delve into the effects of adversarial training on self-supervised and supervised contrastive learning alongside supervised learning. Our analysis uncovers significant disparities between adversarial and clean representations in standard-trained networks across various learning algorithms. Remarkably, adversarial training mitigates these disparities and fosters the convergence of representations toward a universal set, regardless of the learning scheme used. Additionally, increasing the similarity between adversarial and clean representations, particularly near the end of the network, enhances network robustness. These findings offer valuable insights for designing and training effective and robust deep learning networks. Our code is released at \textcolor{magenta}{\url{https://github.com/softsys4ai/CL-Robustness}}.
♻ ☆ FundusQ-Net: a Regression Quality Assessment Deep Learning Algorithm for Fundus Images Quality Grading
Objective: Ophthalmological pathologies such as glaucoma, diabetic retinopathy and age-related macular degeneration are major causes of blindness and vision impairment. There is a need for novel decision support tools that can simplify and speed up the diagnosis of these pathologies. A key step in this process is to automatically estimate the quality of the fundus images to make sure these are interpretable by a human operator or a machine learning model. We present a novel fundus image quality scale and deep learning (DL) model that can estimate fundus image quality relative to this new scale. Methods: A total of 1,245 images were graded for quality by two ophthalmologists within the range 1-10, with a resolution of 0.5. A DL regression model was trained for fundus image quality assessment. The architecture used was Inception-V3. The model was developed using a total of 89,947 images from 6 databases, of which 1,245 were labeled by the specialists and the remaining 88,702 images were used for pre-training and semi-supervised learning. The final DL model was evaluated on an internal test set (n=209) as well as an external test set (n=194). Results: The final DL model, denoted FundusQ-Net, achieved a mean absolute error of 0.61 (0.54-0.68) on the internal test set. When evaluated as a binary classification model on the public DRIMDB database as an external test set the model obtained an accuracy of 99%. Significance: the proposed algorithm provides a new robust tool for automated quality grading of fundus images.
comment: 12 pages, 9 figures, published in Computer Methods and Programs in Biomedicine
♻ ☆ ImageReward: Learning and Evaluating Human Preferences for Text-to-Image Generation
We present a comprehensive solution to learn and improve text-to-image models from human preference feedback. To begin with, we build ImageReward -- the first general-purpose text-to-image human preference reward model -- to effectively encode human preferences. Its training is based on our systematic annotation pipeline including rating and ranking, which collects 137k expert comparisons to date. In human evaluation, ImageReward outperforms existing scoring models and metrics, making it a promising automatic metric for evaluating text-to-image synthesis. On top of it, we propose Reward Feedback Learning (ReFL), a direct tuning algorithm to optimize diffusion models against a scorer. Both automatic and human evaluation support ReFL's advantages over compared methods. All code and datasets are provided at \url{https://github.com/THUDM/ImageReward}.
comment: 32 pages
♻ ☆ EXPRESSNET: An Explainable Residual Slim Network for Fingerprint Presentation Attack Detection
Presentation attack is a challenging issue that persists in the security of automatic fingerprint recognition systems. This paper proposes a novel explainable residual slim network that detects the presentation attack by representing the visual features in the input fingerprint sample. The encoder-decoder of this network along with the channel attention block converts the input sample into its heatmap representation while the modified residual convolutional neural network classifier discriminates between live and spoof fingerprints. The entire architecture of the heatmap generator block and modified ResNet classifier works together in an end-to-end manner. The performance of the proposed model is validated on benchmark liveness detection competition databases i.e. Livdet 2011, 2013, 2015, 2017, and 2019 and the classification accuracy of 96.86\%, 99.84\%, 96.45\%, 96.07\%, 96.27\% are achieved on them, respectively. The performance of the proposed model is compared with the state-of-the-art techniques, and the proposed method outperforms state-of-the-art methods in benchmark protocols of presentation attack detection in terms of classification accuracy.
comment: arXiv admin note: text overlap with arXiv:2303.01465
Information Retrieval 16
☆ On Manipulating Signals of User-Item Graph: A Jacobi Polynomial-based Graph Collaborative Filtering
Collaborative filtering (CF) is an important research direction in recommender systems that aims to make recommendations given the information on user-item interactions. Graph CF has attracted more and more attention in recent years due to its effectiveness in leveraging high-order information in the user-item bipartite graph for better recommendations. Specifically, recent studies show the success of graph neural networks (GNN) for CF is attributed to its low-pass filtering effects. However, current researches lack a study of how different signal components contributes to recommendations, and how to design strategies to properly use them well. To this end, from the view of spectral transformation, we analyze the important factors that a graph filter should consider to achieve better performance. Based on the discoveries, we design JGCF, an efficient and effective method for CF based on Jacobi polynomial bases and frequency decomposition strategies. Extensive experiments on four widely used public datasets show the effectiveness and efficiency of the proposed methods, which brings at most 27.06% performance gain on Alibaba-iFashion. Besides, the experimental results also show that JGCF is better at handling sparse datasets, which shows potential in making recommendations for cold-start users.
☆ Rec4Ad: A Free Lunch to Mitigate Sample Selection Bias for Ads CTR Prediction in Taobao
Click-Through Rate (CTR) prediction serves as a fundamental component in online advertising. A common practice is to train a CTR model on advertisement (ad) impressions with user feedback. Since ad impressions are purposely selected by the model itself, their distribution differs from the inference distribution and thus exhibits sample selection bias (SSB) that affects model performance. Existing studies on SSB mainly employ sample re-weighting techniques which suffer from high variance and poor model calibration. Another line of work relies on costly uniform data that is inadequate to train industrial models. Thus mitigating SSB in industrial models with a uniform-data-free framework is worth exploring. Fortunately, many platforms display mixed results of organic items (i.e., recommendations) and sponsored items (i.e., ads) to users, where impressions of ads and recommendations are selected by different systems but share the same user decision rationales. Based on the above characteristics, we propose to leverage recommendations samples as a free lunch to mitigate SSB for ads CTR model (Rec4Ad). After elaborating data augmentation, Rec4Ad learns disentangled representations with alignment and decorrelation modules for enhancement. When deployed in Taobao display advertising system, Rec4Ad achieves substantial gains in key business metrics, with a lift of up to +6.6\% CTR and +2.9\% RPM.
☆ COPR: Consistency-Oriented Pre-Ranking for Online Advertising
Cascading architecture has been widely adopted in large-scale advertising systems to balance efficiency and effectiveness. In this architecture, the pre-ranking model is expected to be a lightweight approximation of the ranking model, which handles more candidates with strict latency requirements. Due to the gap in model capacity, the pre-ranking and ranking models usually generate inconsistent ranked results, thus hurting the overall system effectiveness. The paradigm of score alignment is proposed to regularize their raw scores to be consistent. However, it suffers from inevitable alignment errors and error amplification by bids when applied in online advertising. To this end, we introduce a consistency-oriented pre-ranking framework for online advertising, which employs a chunk-based sampling module and a plug-and-play rank alignment module to explicitly optimize consistency of ECPM-ranked results. A $\Delta NDCG$-based weighting mechanism is adopted to better distinguish the importance of inter-chunk samples in optimization. Both online and offline experiments have validated the superiority of our framework. When deployed in Taobao display advertising system, it achieves an improvement of up to +12.3\% CTR and +5.6\% RPM.
☆ Generate-then-Retrieve: Intent-Aware FAQ Retrieval in Product Search ACL 2023
Customers interacting with product search engines are increasingly formulating information-seeking queries. Frequently Asked Question (FAQ) retrieval aims to retrieve common question-answer pairs for a user query with question intent. Integrating FAQ retrieval in product search can not only empower users to make more informed purchase decisions, but also enhance user retention through efficient post-purchase support. Determining when an FAQ entry can satisfy a user's information need within product search, without disrupting their shopping experience, represents an important challenge. We propose an intent-aware FAQ retrieval system consisting of (1) an intent classifier that predicts when a user's information need can be answered by an FAQ; (2) a reformulation model that rewrites a query into a natural question. Offline evaluation demonstrates that our approach improves Hit@1 by 13% on retrieving ground-truth FAQs, while reducing latency by 95% compared to baseline systems. These improvements are further validated by real user feedback, where 71% of displayed FAQs on top of product search results received explicit positive user feedback. Overall, our findings show promising directions for integrating FAQ retrieval into product search at scale.
comment: ACL 2023 Industry Track
☆ Computational Technologies for Fashion Recommendation: A Survey
Fashion recommendation is a key research field in computational fashion research and has attracted considerable interest in the computer vision, multimedia, and information retrieval communities in recent years. Due to the great demand for applications, various fashion recommendation tasks, such as personalized fashion product recommendation, complementary (mix-and-match) recommendation, and outfit recommendation, have been posed and explored in the literature. The continuing research attention and advances impel us to look back and in-depth into the field for a better understanding. In this paper, we comprehensively review recent research efforts on fashion recommendation from a technological perspective. We first introduce fashion recommendation at a macro level and analyse its characteristics and differences with general recommendation tasks. We then clearly categorize different fashion recommendation efforts into several sub-tasks and focus on each sub-task in terms of its problem formulation, research focus, state-of-the-art methods, and limitations. We also summarize the datasets proposed in the literature for use in fashion recommendation studies to give readers a brief illustration. Finally, we discuss several promising directions for future research in this field. Overall, this survey systematically reviews the development of fashion recommendation research. It also discusses the current limitations and gaps between academic research and the real needs of the fashion industry. In the process, we offer a deep insight into how the fashion industry could benefit from fashion recommendation technologies. the computational technologies of fashion recommendation.
☆ Tree based Progressive Regression Model for Watch-Time Prediction in Short-video Recommendation
An accurate prediction of watch time has been of vital importance to enhance user engagement in video recommender systems. To achieve this, there are four properties that a watch time prediction framework should satisfy: first, despite its continuous value, watch time is also an ordinal variable and the relative ordering between its values reflects the differences in user preferences. Therefore the ordinal relations should be reflected in watch time predictions. Second, the conditional dependence between the video-watching behaviors should be captured in the model. For instance, one has to watch half of the video before he/she finishes watching the whole video. Third, modeling watch time with a point estimation ignores the fact that models might give results with high uncertainty and this could cause bad cases in recommender systems. Therefore the framework should be aware of prediction uncertainty. Forth, the real-life recommender systems suffer from severe bias amplifications thus an estimation without bias amplification is expected. Therefore we propose TPM for watch time prediction. Specifically, the ordinal ranks of watch time are introduced into TPM and the problem is decomposed into a series of conditional dependent classification tasks which are organized into a tree structure. The expectation of watch time can be generated by traversing the tree and the variance of watch time predictions is explicitly introduced into the objective function as a measurement for uncertainty. Moreover, we illustrate that backdoor adjustment can be seamlessly incorporated into TPM, which alleviates bias amplifications. Extensive offline evaluations have been conducted in public datasets and TPM have been deployed in a real-world video app Kuaishou with over 300 million DAUs. The results indicate that TPM outperforms state-of-the-art approaches and indeed improves video consumption significantly.
☆ Towards Alleviating the Object Bias in Prompt Tuning-based Factual Knowledge Extraction
Many works employed prompt tuning methods to automatically optimize prompt queries and extract the factual knowledge stored in Pretrained Language Models. In this paper, we observe that the optimized prompts, including discrete prompts and continuous prompts, exhibit undesirable object bias. To handle this problem, we propose a novel prompt tuning method called MeCoD. consisting of three modules: Prompt Encoder, Object Equalization and Biased Object Obstruction. Experimental results show that MeCoD can significantly reduce the object bias and at the same time improve accuracy of factual knowledge extraction.
☆ Revisiting Neural Retrieval on Accelerators KDD
Retrieval finds a small number of relevant candidates from a large corpus for information retrieval and recommendation applications. A key component of retrieval is to model (user, item) similarity, which is commonly represented as the dot product of two learned embeddings. This formulation permits efficient inference, commonly known as Maximum Inner Product Search (MIPS). Despite its popularity, dot products cannot capture complex user-item interactions, which are multifaceted and likely high rank. We hence examine non-dot-product retrieval settings on accelerators, and propose \textit{mixture of logits} (MoL), which models (user, item) similarity as an adaptive composition of elementary similarity functions. This new formulation is expressive, capable of modeling high rank (user, item) interactions, and further generalizes to the long tail. When combined with a hierarchical retrieval strategy, \textit{h-indexer}, we are able to scale up MoL to 100M corpus on a single GPU with latency comparable to MIPS baselines. On public datasets, our approach leads to uplifts of up to 77.3\% in hit rate (HR). Experiments on a large recommendation surface at Meta showed strong metric gains and reduced popularity bias, validating the proposed approach's performance and improved generalization.
comment: To appear in the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2023)
☆ Leveraging Explicit Procedural Instructions for Data-Efficient Action Prediction
Task-oriented dialogues often require agents to enact complex, multi-step procedures in order to meet user requests. While large language models have found success automating these dialogues in constrained environments, their widespread deployment is limited by the substantial quantities of task-specific data required for training. The following paper presents a data-efficient solution to constructing dialogue systems, leveraging explicit instructions derived from agent guidelines, such as company policies or customer service manuals. Our proposed Knowledge-Augmented Dialogue System (KADS) combines a large language model with a knowledge retrieval module that pulls documents outlining relevant procedures from a predefined set of policies, given a user-agent interaction. To train this system, we introduce a semi-supervised pre-training scheme that employs dialogue-document matching and action-oriented masked language modeling with partial parameter freezing. We evaluate the effectiveness of our approach on prominent task-oriented dialogue datasets, Action-Based Conversations Dataset and Schema-Guided Dialogue, for two dialogue tasks: action state tracking and workflow discovery. Our results demonstrate that procedural knowledge augmentation improves accuracy predicting in- and out-of-distribution actions while preserving high performance in settings with low or sparse data.
♻ ☆ Criteria Tell You More than Ratings: Criteria Preference-Aware Light Graph Convolution for Effective Multi-Criteria Recommendation KDD
The multi-criteria (MC) recommender system, which leverages MC rating information in a wide range of e-commerce areas, is ubiquitous nowadays. Surprisingly, although graph neural networks (GNNs) have been widely applied to develop various recommender systems due to GNN's high expressive capability in learning graph representations, it has been still unexplored how to design MC recommender systems with GNNs. In light of this, we make the first attempt towards designing a GNN-aided MC recommender system. Specifically, rather than straightforwardly adopting existing GNN-based recommendation methods, we devise a novel criteria preference-aware light graph convolution CPA-LGC method, which is capable of precisely capturing the criteria preference of users as well as the collaborative signal in complex high-order connectivities. To this end, we first construct an MC expansion graph that transforms user--item MC ratings into an expanded bipartite graph to potentially learn from the collaborative signal in MC ratings. Next, to strengthen the capability of criteria preference awareness, CPA-LGC incorporates newly characterized embeddings, including user-specific criteria-preference embeddings and item-specific criterion embeddings, into our graph convolution model. Through comprehensive evaluations using four real-world datasets, we demonstrate (a) the superiority over benchmark MC recommendation methods and benchmark recommendation methods using GNNs with tremendous gains, (b) the effectiveness of core components in CPA-LGC, and (c) the computational efficiency.
comment: 12 pages, 10 figures, 5 tables; 29th ACM SIGKDD Conference on Knowledge Discovery & Data (KDD 2023) (to appear) (Please cite our conference version.)
♻ ☆ The Role of Relevance in Fair Ranking SIGIR 2023
Online platforms mediate access to opportunity: relevance-based rankings create and constrain options by allocating exposure to job openings and job candidates in hiring platforms, or sellers in a marketplace. In order to do so responsibly, these socially consequential systems employ various fairness measures and interventions, many of which seek to allocate exposure based on worthiness. Because these constructs are typically not directly observable, platforms must instead resort to using proxy scores such as relevance and infer them from behavioral signals such as searcher clicks. Yet, it remains an open question whether relevance fulfills its role as such a worthiness score in high-stakes fair rankings. In this paper, we combine perspectives and tools from the social sciences, information retrieval, and fairness in machine learning to derive a set of desired criteria that relevance scores should satisfy in order to meaningfully guide fairness interventions. We then empirically show that not all of these criteria are met in a case study of relevance inferred from biased user click data. We assess the impact of these violations on the estimated system fairness and analyze whether existing fairness interventions may mitigate the identified issues. Our analyses and results surface the pressing need for new approaches to relevance collection and generation that are suitable for use in fair ranking.
comment: Published in SIGIR 2023
♻ ☆ Adap-$τ$: Adaptively Modulating Embedding Magnitude for Recommendation
Recent years have witnessed the great successes of embedding-based methods in recommender systems. Despite their decent performance, we argue one potential limitation of these methods -- the embedding magnitude has not been explicitly modulated, which may aggravate popularity bias and training instability, hindering the model from making a good recommendation. It motivates us to leverage the embedding normalization in recommendation. By normalizing user/item embeddings to a specific value, we empirically observe impressive performance gains (9\% on average) on four real-world datasets. Although encouraging, we also reveal a serious limitation when applying normalization in recommendation -- the performance is highly sensitive to the choice of the temperature $\tau$ which controls the scale of the normalized embeddings. To fully foster the merits of the normalization while circumvent its limitation, this work studied on how to adaptively set the proper $\tau$. Towards this end, we first make a comprehensive analyses of $\tau$ to fully understand its role on recommendation. We then accordingly develop an adaptive fine-grained strategy Adap-$\tau$ for the temperature with satisfying four desirable properties including adaptivity, personalized, efficiency and model-agnostic. Extensive experiments have been conducted to validate the effectiveness of the proposal. The code is available at \url{https://github.com/junkangwu/Adap_tau}.
♻ ☆ Web-Scale Academic Name Disambiguation: the WhoIsWho Benchmark, Leaderboard, and Toolkit KDD 2023
Name disambiguation -- a fundamental problem in online academic systems -- is now facing greater challenges with the increasing growth of research papers. For example, on AMiner, an online academic search platform, about 10% of names own more than 100 authors. Such real-world challenging cases have not been effectively addressed by existing researches due to the small-scale or low-quality datasets that they have used. The development of effective algorithms is further hampered by a variety of tasks and evaluation protocols designed on top of diverse datasets. To this end, we present WhoIsWho owning, a large-scale benchmark with over 1,000,000 papers built using an interactive annotation process, a regular leaderboard with comprehensive tasks, and an easy-to-use toolkit encapsulating the entire pipeline as well as the most powerful features and baseline models for tackling the tasks. Our developed strong baseline has already been deployed online in the AMiner system to enable daily arXiv paper assignments. The public leaderboard is available at http://whoiswho.biendata.xyz/. The toolkit is at https://github.com/THUDM/WhoIsWho. The online demo of daily arXiv paper assignments is at https://na-demo.aminer.cn/arxivpaper.
comment: Accepted by KDD 2023 ADS track
♻ ☆ Influence Maximization with Fairness at Scale (Extended Version)
In this paper, we revisit the problem of influence maximization with fairness, which aims to select k influential nodes to maximise the spread of information in a network, while ensuring that selected sensitive user attributes are fairly affected, i.e., are proportionally similar between the original network and the affected users. Recent studies on this problem focused only on extremely small networks, hence the challenge remains on how to achieve a scalable solution, applicable to networks with millions or billions of nodes. We propose an approach that is based on learning node representations for fair spread from diffusion cascades, instead of the social connectivity s.t. we can deal with very large graphs. We propose two data-driven approaches: (a) fairness-based participant sampling (FPS), and (b) fairness as context (FAC). Spread related user features, such as the probability of diffusing information to others, are derived from the historical information cascades, using a deep neural network. The extracted features are then used in selecting influencers that maximize the influence spread, while being also fair with respect to the chosen sensitive attributes. In FPS, fairness and cascade length information are considered independently in the decision-making process, while FAC considers these information facets jointly and considers correlations between them. The proposed algorithms are generic and represent the first policy-driven solutions that can be applied to arbitrary sets of sensitive attributes at scale. We evaluate the performance of our solutions on a real-world public dataset (Sina Weibo) and on a hybrid real-synthethic dataset (Digg), which exhibit all the facets that we exploit, namely diffusion network, diffusion traces, and user profiles. These experiments show that our methods outperform the state-the-art solutions in terms of spread, fairness, and scalability.
♻ ☆ Generative Flow Network for Listwise Recommendation
Personalized recommender systems fulfill the daily demands of customers and boost online businesses. The goal is to learn a policy that can generate a list of items that matches the user's demand or interest. While most existing methods learn a pointwise scoring model that predicts the ranking score of each individual item, recent research shows that the listwise approach can further improve the recommendation quality by modeling the intra-list correlations of items that are exposed together. This has motivated the recent list reranking and generative recommendation approaches that optimize the overall utility of the entire list. However, it is challenging to explore the combinatorial space of list actions and existing methods that use cross-entropy loss may suffer from low diversity issues. In this work, we aim to learn a policy that can generate sufficiently diverse item lists for users while maintaining high recommendation quality. The proposed solution, GFN4Rec, is a generative method that takes the insight of the flow network to ensure the alignment between list generation probability and its reward. The key advantages of our solution are the log scale reward matching loss that intrinsically improves the generation diversity and the autoregressive item selection model that captures the item mutual influences while capturing future reward of the list. As validation of our method's effectiveness and its superior diversity during active exploration, we conduct experiments on simulated online environments as well as an offline evaluation framework for two real-world datasets.
comment: 11 pages, 5 figures, 9 tables
♻ ☆ Human Mobility Prediction with Causal and Spatial-constrained Multi-task Network
Modeling human mobility helps to understand how people are accessing resources and physically contacting with each other in cities, and thus contributes to various applications such as urban planning, epidemic control, and location-based advertisement. Next location prediction is one decisive task in individual human mobility modeling and is usually viewed as sequence modeling, solved with Markov or RNN-based methods. However, the existing models paid little attention to the logic of individual travel decisions and the reproducibility of the collective behavior of population. To this end, we propose a Causal and Spatial-constrained Long and Short-term Learner (CSLSL) for next location prediction. CSLSL utilizes a causal structure based on multi-task learning to explicitly model the "\textit{when$\rightarrow$what$\rightarrow$where}", a.k.a. "\textit{time$\rightarrow$activity$\rightarrow$location}" decision logic. We next propose a spatial-constrained loss function as an auxiliary task, to ensure the consistency between the predicted and actual spatial distribution of travelers' destinations. Moreover, CSLSL adopts modules named Long and Short-term Capturer (LSC) to learn the transition regularities across different time spans. Extensive experiments on three real-world datasets show promising performance improvements of CSLSL over baselines and confirm the effectiveness of introducing the causality and consistency constraints. The implementation is available at https://github.com/urbanmobility/CSLSL.
comment: Updated version (We have corrected the title error in the previous version.)
Machine Learning 150
☆ ChatDB: Augmenting LLMs with Databases as Their Symbolic Memory
Large language models (LLMs) with memory are computationally universal. However, mainstream LLMs are not taking full advantage of memory, and the designs are heavily influenced by biological brains. Due to their approximate nature and proneness to the accumulation of errors, conventional neural memory mechanisms cannot support LLMs to simulate complex reasoning. In this paper, we seek inspiration from modern computer architectures to augment LLMs with symbolic memory for complex multi-hop reasoning. Such a symbolic memory framework is instantiated as an LLM and a set of SQL databases, where the LLM generates SQL instructions to manipulate the SQL databases. We validate the effectiveness of the proposed memory framework on a synthetic dataset requiring complex reasoning. The project website is available at https://chatdatabase.github.io/ .
☆ Model Spider: Learning to Rank Pre-Trained Models Efficiently
Figuring out which Pre-Trained Model (PTM) from a model zoo fits the target task is essential to take advantage of plentiful model resources. With the availability of numerous heterogeneous PTMs from diverse fields, efficiently selecting the most suitable PTM is challenging due to the time-consuming costs of carrying out forward or backward passes over all PTMs. In this paper, we propose Model Spider, which tokenizes both PTMs and tasks by summarizing their characteristics into vectors to enable efficient PTM selection. By leveraging the approximated performance of PTMs on a separate set of training tasks, Model Spider learns to construct tokens and measure the fitness score between a model-task pair via their tokens. The ability to rank relevant PTMs higher than others generalizes to new tasks. With the top-ranked PTM candidates, we further learn to enrich task tokens with their PTM-specific semantics to re-rank the PTMs for better selection. Model Spider balances efficiency and selection ability, making PTM selection like a spider preying on a web. Model Spider demonstrates promising performance in various configurations of model zoos.
☆ Fast Context Adaptation in Cost-Aware Continual Learning
In the past few years, DRL has become a valuable solution to automatically learn efficient resource management strategies in complex networks with time-varying statistics. However, the increased complexity of 5G and Beyond networks requires correspondingly more complex learning agents and the learning process itself might end up competing with users for communication and computational resources. This creates friction: on the one hand, the learning process needs resources to quickly convergence to an effective strategy; on the other hand, the learning process needs to be efficient, i.e., take as few resources as possible from the user's data plane, so as not to throttle users' QoS. In this paper, we investigate this trade-off and propose a dynamic strategy to balance the resources assigned to the data plane and those reserved for learning. With the proposed approach, a learning agent can quickly converge to an efficient resource allocation strategy and adapt to changes in the environment as for the CL paradigm, while minimizing the impact on the users' QoS. Simulation results show that the proposed method outperforms static allocation methods with minimal learning overhead, almost reaching the performance of an ideal out-of-band CL solution.
comment: arXiv admin note: text overlap with arXiv:2211.16915
☆ Deductive Verification of Chain-of-Thought Reasoning
Large Language Models (LLMs) significantly benefit from Chain-of-Thought (CoT) prompting in performing various reasoning tasks. While CoT allows models to produce more comprehensive reasoning processes, its emphasis on intermediate reasoning steps can inadvertently introduce hallucinations and accumulated errors, thereby limiting models' ability to solve complex reasoning tasks. Inspired by how humans engage in careful and meticulous deductive logical reasoning processes to solve tasks, we seek to enable language models to perform explicit and rigorous deductive reasoning, and also ensure the trustworthiness of their reasoning process through self-verification. However, directly verifying the validity of an entire deductive reasoning process is challenging, even with advanced models like ChatGPT. In light of this, we propose to decompose a reasoning verification process into a series of step-by-step subprocesses, each only receiving their necessary context and premises. To facilitate this procedure, we propose Natural Program, a natural language-based deductive reasoning format. Our approach enables models to generate precise reasoning steps where subsequent steps are more rigorously grounded on prior steps. It also empowers language models to carry out reasoning self-verification in a step-by-step manner. By integrating this verification process into each deductive reasoning stage, we significantly enhance the rigor and trustfulness of generated reasoning steps. Along this process, we also improve the answer correctness on complex reasoning tasks. Code will be released at https://github.com/lz1oceani/verify_cot.
☆ Spherical Fourier Neural Operators: Learning Stable Dynamics on the Sphere
Fourier Neural Operators (FNOs) have proven to be an efficient and effective method for resolution-independent operator learning in a broad variety of application areas across scientific machine learning. A key reason for their success is their ability to accurately model long-range dependencies in spatio-temporal data by learning global convolutions in a computationally efficient manner. To this end, FNOs rely on the discrete Fourier transform (DFT), however, DFTs cause visual and spectral artifacts as well as pronounced dissipation when learning operators in spherical coordinates since they incorrectly assume a flat geometry. To overcome this limitation, we generalize FNOs on the sphere, introducing Spherical FNOs (SFNOs) for learning operators on spherical geometries. We apply SFNOs to forecasting atmospheric dynamics, and demonstrate stable auto\-regressive rollouts for a year of simulated time (1,460 steps), while retaining physically plausible dynamics. The SFNO has important implications for machine learning-based simulation of climate dynamics that could eventually help accelerate our response to climate change.
☆ Atrial Septal Defect Detection in Children Based on Ultrasound Video Using Multiple Instances Learning
Purpose: Congenital heart defect (CHD) is the most common birth defect. Thoracic echocardiography (TTE) can provide sufficient cardiac structure information, evaluate hemodynamics and cardiac function, and is an effective method for atrial septal defect (ASD) examination. This paper aims to study a deep learning method based on cardiac ultrasound video to assist in ASD diagnosis. Materials and methods: We select two standard views of the atrial septum (subAS) and low parasternal four-compartment view (LPS4C) as the two views to identify ASD. We enlist data from 300 children patients as part of a double-blind experiment for five-fold cross-validation to verify the performance of our model. In addition, data from 30 children patients (15 positives and 15 negatives) are collected for clinician testing and compared to our model test results (these 30 samples do not participate in model training). We propose an echocardiography video-based atrial septal defect diagnosis system. In our model, we present a block random selection, maximal agreement decision and frame sampling strategy for training and testing respectively, resNet18 and r3D networks are used to extract the frame features and aggregate them to build a rich video-level representation. Results: We validate our model using our private dataset by five-cross validation. For ASD detection, we achieve 89.33 AUC, 84.95 accuracy, 85.70 sensitivity, 81.51 specificity and 81.99 F1 score. Conclusion: The proposed model is multiple instances learning-based deep learning model for video atrial septal defect detection which effectively improves ASD detection accuracy when compared to the performances of previous networks and clinical doctors.
☆ MTS2Graph: Interpretable Multivariate Time Series Classification with Temporal Evolving Graphs
Conventional time series classification approaches based on bags of patterns or shapelets face significant challenges in dealing with a vast amount of feature candidates from high-dimensional multivariate data. In contrast, deep neural networks can learn low-dimensional features efficiently, and in particular, Convolutional Neural Networks (CNN) have shown promising results in classifying Multivariate Time Series (MTS) data. A key factor in the success of deep neural networks is this astonishing expressive power. However, this power comes at the cost of complex, black-boxed models, conflicting with the goals of building reliable and human-understandable models. An essential criterion in understanding such predictive deep models involves quantifying the contribution of time-varying input variables to the classification. Hence, in this work, we introduce a new framework for interpreting multivariate time series data by extracting and clustering the input representative patterns that highly activate CNN neurons. This way, we identify each signal's role and dependencies, considering all possible combinations of signals in the MTS input. Then, we construct a graph that captures the temporal relationship between the extracted patterns for each layer. An effective graph merging strategy finds the connection of each node to the previous layer's nodes. Finally, a graph embedding algorithm generates new representations of the created interpretable time-series features. To evaluate the performance of our proposed framework, we run extensive experiments on eight datasets of the UCR/UEA archive, along with HAR and PAM datasets. The experiments indicate the benefit of our time-aware graph-based representation in MTS classification while enriching them with more interpretability.
☆ Patient Dropout Prediction in Virtual Health: A Multimodal Dynamic Knowledge Graph and Text Mining Approach
Virtual health has been acclaimed as a transformative force in healthcare delivery. Yet, its dropout issue is critical that leads to poor health outcomes, increased health, societal, and economic costs. Timely prediction of patient dropout enables stakeholders to take proactive steps to address patients' concerns, potentially improving retention rates. In virtual health, the information asymmetries inherent in its delivery format, between different stakeholders, and across different healthcare delivery systems hinder the performance of existing predictive methods. To resolve those information asymmetries, we propose a Multimodal Dynamic Knowledge-driven Dropout Prediction (MDKDP) framework that learns implicit and explicit knowledge from doctor-patient dialogues and the dynamic and complex networks of various stakeholders in both online and offline healthcare delivery systems. We evaluate MDKDP by partnering with one of the largest virtual health platforms in China. MDKDP improves the F1-score by 3.26 percentage points relative to the best benchmark. Comprehensive robustness analyses show that integrating stakeholder attributes, knowledge dynamics, and compact bilinear pooling significantly improves the performance. Our work provides significant implications for healthcare IT by revealing the value of mining relations and knowledge across different service modalities. Practically, MDKDP offers a novel design artifact for virtual health platforms in patient dropout management.
☆ Sequential Principal-Agent Problems with Communication: Efficient Computation and Learning
We study a sequential decision making problem between a principal and an agent with incomplete information on both sides. In this model, the principal and the agent interact in a stochastic environment, and each is privy to observations about the state not available to the other. The principal has the power of commitment, both to elicit information from the agent and to provide signals about her own information. The principal and the agent communicate their signals to each other, and select their actions independently based on this communication. Each player receives a payoff based on the state and their joint actions, and the environment moves to a new state. The interaction continues over a finite time horizon, and both players act to optimize their own total payoffs over the horizon. Our model encompasses as special cases stochastic games of incomplete information and POMDPs, as well as sequential Bayesian persuasion and mechanism design problems. We study both computation of optimal policies and learning in our setting. While the general problems are computationally intractable, we study algorithmic solutions under a conditional independence assumption on the underlying state-observation distributions. We present an polynomial-time algorithm to compute the principal's optimal policy up to an additive approximation. Additionally, we show an efficient learning algorithm in the case where the transition probabilities are not known beforehand. The algorithm guarantees sublinear regret for both players.
GEO-Bench: Toward Foundation Models for Earth Monitoring
Recent progress in self-supervision has shown that pre-training large neural networks on vast amounts of unsupervised data can lead to substantial increases in generalization to downstream tasks. Such models, recently coined foundation models, have been transformational to the field of natural language processing. Variants have also been proposed for image data, but their applicability to remote sensing tasks is limited. To stimulate the development of foundation models for Earth monitoring, we propose a benchmark comprised of six classification and six segmentation tasks, which were carefully curated and adapted to be both relevant to the field and well-suited for model evaluation. We accompany this benchmark with a robust methodology for evaluating models and reporting aggregated results to enable a reliable assessment of progress. Finally, we report results for 20 baselines to gain information about the performance of existing models. We believe that this benchmark will be a driver of progress across a variety of Earth monitoring tasks.
☆ Inductive Bias for Emergent Communication in a Continuous Setting NIPS 2023
We study emergent communication in a multi-agent reinforcement learning setting, where the agents solve cooperative tasks and have access to a communication channel. The communication channel may consist of either discrete symbols or continuous variables. We introduce an inductive bias to aid with the emergence of good communication protocols for continuous messages, and we look at the effect this type of inductive bias has for continuous and discrete messages in itself or when used in combination with reinforcement learning. We demonstrate that this type of inductive bias has a beneficial effect on the communication protocols learnt in two toy environments, Negotiation and Sequence Guess.
comment: NIPS 2023 Preprint. 12 pages, 5 figures, 3 tables
☆ Quick-Tune: Quickly Learning Which Pretrained Model to Finetune and How
With the ever-increasing number of pretrained models, machine learning practitioners are continuously faced with which pretrained model to use, and how to finetune it for a new dataset. In this paper, we propose a methodology that jointly searches for the optimal pretrained model and the hyperparameters for finetuning it. Our method transfers knowledge about the performance of many pretrained models with multiple hyperparameter configurations on a series of datasets. To this aim, we evaluated over 20k hyperparameter configurations for finetuning 24 pretrained image classification models on 87 datasets to generate a large-scale meta-dataset. We meta-learn a multi-fidelity performance predictor on the learning curves of this meta-dataset and use it for fast hyperparameter optimization on new datasets. We empirically demonstrate that our resulting approach can quickly select an accurate pretrained model for a new dataset together with its optimal hyperparameters.
☆ Understanding Generalization of Federated Learning via Stability: Heterogeneity Matters NeurIPS 2023
Generalization performance is a key metric in evaluating machine learning models when applied to real-world applications. Good generalization indicates the model can predict unseen data correctly when trained under a limited number of data. Federated learning (FL), which has emerged as a popular distributed learning framework, allows multiple devices or clients to train a shared model without violating privacy requirements. While the existing literature has studied extensively the generalization performances of centralized machine learning algorithms, similar analysis in the federated settings is either absent or with very restrictive assumptions on the loss functions. In this paper, we aim to analyze the generalization performances of federated learning by means of algorithmic stability, which measures the change of the output model of an algorithm when perturbing one data point. Three widely-used algorithms are studied, including FedAvg, SCAFFOLD, and FedProx, under convex and non-convex loss functions. Our analysis shows that the generalization performances of models trained by these three algorithms are closely related to the heterogeneity of clients' datasets as well as the convergence behaviors of the algorithms. Particularly, in the i.i.d. setting, our results recover the classical results of stochastic gradient descent (SGD).
comment: Submitted to NeurIPS 2023
☆ LEACE: Perfect linear concept erasure in closed form
Concept erasure aims to remove specified features from a representation. It can be used to improve fairness (e.g. preventing a classifier from using gender or race) and interpretability (e.g. removing a concept to observe changes in model behavior). In this paper, we introduce LEAst-squares Concept Erasure (LEACE), a closed-form method which provably prevents all linear classifiers from detecting a concept while inflicting the least possible damage to the representation. We apply LEACE to large language models with a novel procedure called "concept scrubbing," which erases target concept information from every layer in the network. We demonstrate the usefulness of our method on two tasks: measuring the reliance of language models on part-of-speech information, and reducing gender bias in BERT embeddings. Code is available at https://github.com/EleutherAI/concept-erasure.
☆ Computation with Sequences in the Brain
Even as machine learning exceeds human-level performance on many applications, the generality, robustness, and rapidity of the brain's learning capabilities remain unmatched. How cognition arises from neural activity is a central open question in neuroscience, inextricable from the study of intelligence itself. A simple formal model of neural activity was proposed in Papadimitriou [2020] and has been subsequently shown, through both mathematical proofs and simulations, to be capable of implementing certain simple cognitive operations via the creation and manipulation of assemblies of neurons. However, many intelligent behaviors rely on the ability to recognize, store, and manipulate temporal sequences of stimuli (planning, language, navigation, to list a few). Here we show that, in the same model, time can be captured naturally as precedence through synaptic weights and plasticity, and, as a result, a range of computations on sequences of assemblies can be carried out. In particular, repeated presentation of a sequence of stimuli leads to the memorization of the sequence through corresponding neural assemblies: upon future presentation of any stimulus in the sequence, the corresponding assembly and its subsequent ones will be activated, one after the other, until the end of the sequence. Finally, we show that any finite state machine can be learned in a similar way, through the presentation of appropriate patterns of sequences. Through an extension of this mechanism, the model can be shown to be capable of universal computation. We support our analysis with a number of experiments to probe the limits of learning in this model in key ways. Taken together, these results provide a concrete hypothesis for the basis of the brain's remarkable abilities to compute and learn, with sequences playing a vital role.
comment: 24 pages, 12 figures
☆ The Emergence of Essential Sparsity in Large Pre-trained Models: The Weights that Matter
Large pre-trained transformers are show-stealer in modern-day deep learning, and it becomes crucial to comprehend the parsimonious patterns that exist within them as they grow in scale. With exploding parameter counts, Lottery Ticket Hypothesis (LTH) and its variants, have lost their pragmatism in sparsifying them due to high computation and memory bottleneck of the repetitive train-prune-retrain routine of iterative magnitude pruning (IMP) which worsens with increasing model size. In this paper, we comprehensively study induced sparse patterns across multiple large pre-trained vision and language transformers. We propose the existence of -- essential sparsity defined with a sharp dropping point beyond which the performance declines much faster w.r.t the rise of sparsity level, when we directly remove weights with the smallest magnitudes in one-shot. In the sparsity-performance curve We also present an intriguing emerging phenomenon of abrupt sparsification during the pre-training of BERT, i.e., BERT suddenly becomes heavily sparse in pre-training after certain iterations. Moreover, our observations also indicate a counter-intuitive finding that BERT trained with a larger amount of pre-training data tends to have a better ability to condense knowledge in comparatively relatively fewer parameters. Lastly, we investigate the effect of the pre-training loss on essential sparsity and discover that self-supervised learning (SSL) objectives trigger stronger emergent sparsification properties than supervised learning (SL). Our codes are available at \url{https://github.com/VITA-Group/essential\_sparsity}.
☆ Stable Vectorization of Multiparameter Persistent Homology using Signed Barcodes as Measures
Persistent homology (PH) provides topological descriptors for geometric data, such as weighted graphs, which are interpretable, stable to perturbations, and invariant under, e.g., relabeling. Most applications of PH focus on the one-parameter case -- where the descriptors summarize the changes in topology of data as it is filtered by a single quantity of interest -- and there is now a wide array of methods enabling the use of one-parameter PH descriptors in data science, which rely on the stable vectorization of these descriptors as elements of a Hilbert space. Although the multiparameter PH (MPH) of data that is filtered by several quantities of interest encodes much richer information than its one-parameter counterpart, the scarceness of stability results for MPH descriptors has so far limited the available options for the stable vectorization of MPH. In this paper, we aim to bring together the best of both worlds by showing how the interpretation of signed barcodes -- a recent family of MPH descriptors -- as signed measures leads to natural extensions of vectorization strategies from one parameter to multiple parameters. The resulting feature vectors are easy to define and to compute, and provably stable. While, as a proof of concept, we focus on simple choices of signed barcodes and vectorizations, we already see notable performance improvements when comparing our feature vectors to state-of-the-art topology-based methods on various types of data.
comment: 23 pages, 3 figures, 8 tables
☆ AI-Supported Assessment of Load Safety
Load safety assessment and compliance is an essential step in the corporate process of every logistics service provider. In 2020, a total of 11,371 police checks of trucks were carried out, during which 9.6% (1091) violations against the load safety regulations were detected. For a logistic service provider, every load safety violation results in height fines and damage to reputation. An assessment of load safety supported by artificial intelligence (AI) will reduce the risk of accidents by unsecured loads and fines during safety assessments. This work shows how photos of the load, taken by the truck driver or the loadmaster after the loading process, can be used to assess load safety. By a trained two-stage artificial neural network (ANN), these photos are classified into three different classes I) cargo loaded safely, II) cargo loaded unsafely, and III) unusable image. By applying several architectures of convolutional neural networks (CNN), it can be shown that it is possible to distinguish between unusable and usable images for cargo safety assessment. This distinction is quite crucial since the truck driver and the loadmaster sometimes provide photos without the essential image features like the case structure of the truck and the whole cargo. A human operator or another ANN will then assess the load safety within the second stage.
comment: 9 pages, 4 figures, 2 tables
☆ FAMO: Fast Adaptive Multitask Optimization
One of the grand enduring goals of AI is to create generalist agents that can learn multiple different tasks from diverse data via multitask learning (MTL). However, gradient descent (GD) on the average loss across all tasks may yield poor multitask performance due to severe under-optimization of certain tasks. Previous approaches that manipulate task gradients for a more balanced loss decrease require storing and computing all task gradients (O(K) space and time where K is the number of tasks), limiting their use in large-scale scenarios. In this work, we introduce Fast Adaptive Multitask Optimization (FAMO), a dynamic weighting method that decreases task losses in a balanced way using O(1) space and time. We conduct an extensive set of experiments covering multi-task supervised and reinforcement learning problems. Our results indicate that FAMO achieves comparable or superior performance to state-of-the-art gradient manipulation techniques while offering significant improvements in space and computational efficiency. Code is available at https://github.com/Cranial-XIX/FAMO.
☆ Asymptotics of Bayesian Uncertainty Estimation in Random Features Regression
In this paper we compare and contrast the behavior of the posterior predictive distribution to the risk of the maximum a posteriori estimator for the random features regression model in the overparameterized regime. We will focus on the variance of the posterior predictive distribution (Bayesian model average) and compare its asymptotics to that of the risk of the MAP estimator. In the regime where the model dimensions grow faster than any constant multiple of the number of samples, asymptotic agreement between these two quantities is governed by the phase transition in the signal-to-noise ratio. They also asymptotically agree with each other when the number of samples grow faster than any constant multiple of model dimensions. Numerical simulations illustrate finer distributional properties of the two quantities for finite dimensions. We conjecture they have Gaussian fluctuations and exhibit similar properties as found by previous authors in a Gaussian sequence model, which is of independent theoretical interest.
comment: 11 pages, 3 figures
☆ Matched Pair Calibration for Ranking Fairness
We propose a test of fairness in score-based ranking systems called matched pair calibration. Our approach constructs a set of matched item pairs with minimal confounding differences between subgroups before computing an appropriate measure of ranking error over the set. The matching step ensures that we compare subgroup outcomes between identically scored items so that measured performance differences directly imply unfairness in subgroup-level exposures. We show how our approach generalizes the fairness intuitions of calibration from a binary classification setting to ranking and connect our approach to other proposals for ranking fairness measures. Moreover, our strategy shows how the logic of marginal outcome tests extends to cases where the analyst has access to model scores. Lastly, we provide an example of applying matched pair calibration to a real-word ranking data set to demonstrate its efficacy in detecting ranking bias.
comment: 19 pages, 8 figures
☆ Graph Classification Gaussian Processes via Spectral Features
Graph classification aims to categorise graphs based on their structure and node attributes. In this work, we propose to tackle this task using tools from graph signal processing by deriving spectral features, which we then use to design two variants of Gaussian process models for graph classification. The first variant uses spectral features based on the distribution of energy of a node feature signal over the spectrum of the graph. We show that even such a simple approach, having no learned parameters, can yield competitive performance compared to strong neural network and graph kernel baselines. A second, more sophisticated variant is designed to capture multi-scale and localised patterns in the graph by learning spectral graph wavelet filters, obtaining improved performance on synthetic and real-world data sets. Finally, we show that both models produce well calibrated uncertainty estimates, enabling reliable decision making based on the model predictions.
☆ Exploring the effects of robotic design on learning and neural control
The ongoing deep learning revolution has allowed computers to outclass humans in various games and perceive features imperceptible to humans during classification tasks. Current machine learning techniques have clearly distinguished themselves in specialized tasks. However, we have yet to see robots capable of performing multiple tasks at an expert level. Most work in this field is focused on the development of more sophisticated learning algorithms for a robot's controller given a largely static and presupposed robotic design. By focusing on the development of robotic bodies, rather than neural controllers, I have discovered that robots can be designed such that they overcome many of the current pitfalls encountered by neural controllers in multitask settings. Through this discovery, I also present novel metrics to explicitly measure the learning ability of a robotic design and its resistance to common problems such as catastrophic interference. Traditionally, the physical robot design requires human engineers to plan every aspect of the system, which is expensive and often relies on human intuition. In contrast, within the field of evolutionary robotics, evolutionary algorithms are used to automatically create optimized designs, however, such designs are often still limited in their ability to perform in a multitask setting. The metrics created and presented here give a novel path to automated design that allow evolved robots to synergize with their controller to improve the computational efficiency of their learning while overcoming catastrophic interference. Overall, this dissertation intimates the ability to automatically design robots that are more general purpose than current robots and that can perform various tasks while requiring less computation.
comment: arXiv admin note: text overlap with arXiv:2008.06397
☆ Soft Merging of Experts with Adaptive Routing
Sparsely activated neural networks with conditional computation learn to route their inputs through different "expert" subnetworks, providing a form of modularity that densely activated models lack. Despite their possible benefits, models with learned routing often underperform their parameter-matched densely activated counterparts as well as models that use non-learned heuristic routing strategies. In this paper, we hypothesize that these shortcomings stem from the gradient estimation techniques used to train sparsely activated models that use non-differentiable discrete routing decisions. To address this issue, we introduce Soft Merging of Experts with Adaptive Routing (SMEAR), which avoids discrete routing by using a single "merged" expert constructed via a weighted average of all of the experts' parameters. By routing activations through a single merged expert, SMEAR does not incur a significant increase in computational costs and enables standard gradient-based training. We empirically validate that models using SMEAR outperform models that route based on metadata or learn sparse routing through gradient estimation. Furthermore, we provide qualitative analysis demonstrating that the experts learned via SMEAR exhibit a significant amount of specialization. All of the code used in our experiments is publicly available.
☆ Learning to Do or Learning While Doing: Reinforcement Learning and Bayesian Optimisation for Online Continuous Tuning
Online tuning of real-world plants is a complex optimisation problem that continues to require manual intervention by experienced human operators. Autonomous tuning is a rapidly expanding field of research, where learning-based methods, such as Reinforcement Learning-trained Optimisation (RLO) and Bayesian optimisation (BO), hold great promise for achieving outstanding plant performance and reducing tuning times. Which algorithm to choose in different scenarios, however, remains an open question. Here we present a comparative study using a routine task in a real particle accelerator as an example, showing that RLO generally outperforms BO, but is not always the best choice. Based on the study's results, we provide a clear set of criteria to guide the choice of algorithm for a given tuning task. These can ease the adoption of learning-based autonomous tuning solutions to the operation of complex real-world plants, ultimately improving the availability and pushing the limits of operability of these facilities, thereby enabling scientific and engineering advancements.
comment: 17 pages, 8 figures, 2 tables
☆ Towards Visual Foundational Models of Physical Scenes
We describe a first step towards learning general-purpose visual representations of physical scenes using only image prediction as a training criterion. To do so, we first define "physical scene" and show that, even though different agents may maintain different representations of the same scene, the underlying physical scene that can be inferred is unique. Then, we show that NeRFs cannot represent the physical scene, as they lack extrapolation mechanisms. Those, however, could be provided by Diffusion Models, at least in theory. To test this hypothesis empirically, NeRFs can be combined with Diffusion Models, a process we refer to as NeRF Diffusion, used as unsupervised representations of the physical scene. Our analysis is limited to visual data, without external grounding mechanisms that can be provided by independent sensory modalities.
comment: TLDR: Physical scenes are equivalence classes of sufficient statistics, and can be inferred uniquely by any agent measuring the same finite data; We formalize and implement an approach to representation learning that overturns "naive realism" in favor of an analytical approach of Russell and Koenderink. NeRFs cannot capture the physical scenes, but combined with Diffusion Models they can
☆ Exploring Model Dynamics for Accumulative Poisoning Discovery ICML 2023
Adversarial poisoning attacks pose huge threats to various machine learning applications. Especially, the recent accumulative poisoning attacks show that it is possible to achieve irreparable harm on models via a sequence of imperceptible attacks followed by a trigger batch. Due to the limited data-level discrepancy in real-time data streaming, current defensive methods are indiscriminate in handling the poison and clean samples. In this paper, we dive into the perspective of model dynamics and propose a novel information measure, namely, Memorization Discrepancy, to explore the defense via the model-level information. By implicitly transferring the changes in the data manipulation to that in the model outputs, Memorization Discrepancy can discover the imperceptible poison samples based on their distinct dynamics from the clean samples. We thoroughly explore its properties and propose Discrepancy-aware Sample Correction (DSC) to defend against accumulative poisoning attacks. Extensive experiments comprehensively characterized Memorization Discrepancy and verified its effectiveness. The code is publicly available at: https://github.com/tmlr-group/Memorization-Discrepancy.
comment: accepted by ICML 2023
☆ Towards Memory-Efficient Training for Extremely Large Output Spaces -- Learning with 500k Labels on a Single Commodity GPU
In classification problems with large output spaces (up to millions of labels), the last layer can require an enormous amount of memory. Using sparse connectivity would drastically reduce the memory requirements, but as we show below, it can result in much diminished predictive performance of the model. Fortunately, we found that this can be mitigated by introducing a penultimate layer of intermediate size. We further demonstrate that one can constrain the connectivity of the sparse layer to be uniform, in the sense that each output neuron will have the exact same number of incoming connections. This allows for efficient implementations of sparse matrix multiplication and connection redistribution on GPU hardware. Via a custom CUDA implementation, we show that the proposed approach can scale to datasets with 670,000 labels on a single commodity GPU with only 4GB memory.
☆ Emotion-Conditioned Melody Harmonization with Hierarchical Variational Autoencoder
Existing melody harmonization models have made great progress in improving the quality of generated harmonies, but most of them ignored the emotions beneath the music. Meanwhile, the variability of harmonies generated by previous methods is insufficient. To solve these problems, we propose a novel LSTM-based Hierarchical Variational Auto-Encoder (LHVAE) to investigate the influence of emotional conditions on melody harmonization, while improving the quality of generated harmonies and capturing the abundant variability of chord progressions. Specifically, LHVAE incorporates latent variables and emotional conditions at different levels (piece- and bar-level) to model the global and local music properties. Additionally, we introduce an attention-based melody context vector at each step to better learn the correspondence between melodies and harmonies. Experimental results of the objective evaluation show that our proposed model outperforms other LSTM-based models. Through subjective evaluation, we conclude that only altering the chords hardly changes the overall emotion of the music. The qualitative analysis demonstrates the ability of our model to generate variable harmonies.
comment: Accepted by IEEE SMC 2023
☆ Unleashing Mask: Explore the Intrinsic Out-of-Distribution Detection Capability ICML 2023
Out-of-distribution (OOD) detection is an indispensable aspect of secure AI when deploying machine learning models in real-world applications. Previous paradigms either explore better scoring functions or utilize the knowledge of outliers to equip the models with the ability of OOD detection. However, few of them pay attention to the intrinsic OOD detection capability of the given model. In this work, we generally discover the existence of an intermediate stage of a model trained on in-distribution (ID) data having higher OOD detection performance than that of its final stage across different settings, and further identify one critical data-level attribution to be learning with the atypical samples. Based on such insights, we propose a novel method, Unleashing Mask, which aims to restore the OOD discriminative capabilities of the well-trained model with ID data. Our method utilizes a mask to figure out the memorized atypical samples, and then finetune the model or prune it with the introduced mask to forget them. Extensive experiments and analysis demonstrate the effectiveness of our method. The code is available at: https://github.com/tmlr-group/Unleashing-Mask.
comment: accepted by ICML 2023
☆ Bayesian post-hoc regularization of random forests
Random Forests are powerful ensemble learning algorithms widely used in various machine learning tasks. However, they have a tendency to overfit noisy or irrelevant features, which can result in decreased generalization performance. Post-hoc regularization techniques aim to mitigate this issue by modifying the structure of the learned ensemble after its training. Here, we propose Bayesian post-hoc regularization to leverage the reliable patterns captured by leaf nodes closer to the root, while potentially reducing the impact of more specific and potentially noisy leaf nodes deeper in the tree. This approach allows for a form of pruning that does not alter the general structure of the trees but rather adjusts the influence of leaf nodes based on their proximity to the root node. We have evaluated the performance of our method on various machine learning data sets. Our approach demonstrates competitive performance with the state-of-the-art methods and, in certain cases, surpasses them in terms of predictive accuracy and generalization.
☆ Fine-grained Expressivity of Graph Neural Networks
Numerous recent works have analyzed the expressive power of message-passing graph neural networks (MPNNs), primarily utilizing combinatorial techniques such as the $1$-dimensional Weisfeiler-Leman test ($1$-WL) for the graph isomorphism problem. However, the graph isomorphism objective is inherently binary, not giving insights into the degree of similarity between two given graphs. This work resolves this issue by considering continuous extensions of both $1$-WL and MPNNs to graphons. Concretely, we show that the continuous variant of $1$-WL delivers an accurate topological characterization of the expressive power of MPNNs on graphons, revealing which graphs these networks can distinguish and the level of difficulty in separating them. We identify the finest topology where MPNNs separate points and prove a universal approximation theorem. Consequently, we provide a theoretical framework for graph and graphon similarity combining various topological variants of classical characterizations of the $1$-WL. In particular, we characterize the expressive power of MPNNs in terms of the tree distance, which is a graph distance based on the concepts of fractional isomorphisms, and substructure counts via tree homomorphisms, showing that these concepts have the same expressive power as the $1$-WL and MPNNs on graphons. Empirically, we validate our theoretical findings by showing that randomly initialized MPNNs, without training, exhibit competitive performance compared to their trained counterparts. Moreover, we evaluate different MPNN architectures based on their ability to preserve graph distances, highlighting the significance of our continuous $1$-WL test in understanding MPNNs' expressivity.
☆ Mildly Constrained Evaluation Policy for Offline Reinforcement Learning
Offline reinforcement learning (RL) methodologies enforce constraints on the policy to adhere closely to the behavior policy, thereby stabilizing value learning and mitigating the selection of out-of-distribution (OOD) actions during test time. Conventional approaches apply identical constraints for both value learning and test time inference. However, our findings indicate that the constraints suitable for value estimation may in fact be excessively restrictive for action selection during test time. To address this issue, we propose a Mildly Constrained Evaluation Policy (MCEP) for test time inference with a more constrained target policy for value estimation. Since the target policy has been adopted in various prior approaches, MCEP can be seamlessly integrated with them as a plug-in. We instantiate MCEP based on TD3-BC [Fujimoto and Gu, 2021] and AWAC [Nair et al., 2020] algorithms. The empirical results on MuJoCo locomotion tasks show that the MCEP significantly outperforms the target policy and achieves competitive results to state-of-the-art offline RL methods. The codes are open-sourced at https://github.com/egg-west/MCEP.git.
☆ Human-imperceptible, Machine-recognizable Images
Massive human-related data is collected to train neural networks for computer vision tasks. A major conflict is exposed relating to software engineers between better developing AI systems and distancing from the sensitive training data. To reconcile this conflict, this paper proposes an efficient privacy-preserving learning paradigm, where images are first encrypted to become ``human-imperceptible, machine-recognizable'' via one of the two encryption strategies: (1) random shuffling to a set of equally-sized patches and (2) mixing-up sub-patches of the images. Then, minimal adaptations are made to vision transformer to enable it to learn on the encrypted images for vision tasks, including image classification and object detection. Extensive experiments on ImageNet and COCO show that the proposed paradigm achieves comparable accuracy with the competitive methods. Decrypting the encrypted images requires solving an NP-hard jigsaw puzzle or an ill-posed inverse problem, which is empirically shown intractable to be recovered by various attackers, including the powerful vision transformer-based attacker. We thus show that the proposed paradigm can ensure the encrypted images have become human-imperceptible while preserving machine-recognizable information. The code is available at \url{https://github.com/FushengHao/PrivacyPreservingML.}
☆ Schema First! Learn Versatile Knowledge Graph Embeddings by Capturing Semantics with MASCHInE
Knowledge graph embedding models (KGEMs) have gained considerable traction in recent years. These models learn a vector representation of knowledge graph entities and relations, a.k.a. knowledge graph embeddings (KGEs). Learning versatile KGEs is desirable as it makes them useful for a broad range of tasks. However, KGEMs are usually trained for a specific task, which makes their embeddings task-dependent. In parallel, the widespread assumption that KGEMs actually create a semantic representation of the underlying entities and relations (e.g., project similar entities closer than dissimilar ones) has been challenged. In this work, we design heuristics for generating protographs -- small, modified versions of a KG that leverage schema-based information. The learnt protograph-based embeddings are meant to encapsulate the semantics of a KG, and can be leveraged in learning KGEs that, in turn, also better capture semantics. Extensive experiments on various evaluation benchmarks demonstrate the soundness of this approach, which we call Modular and Agnostic SCHema-based Integration of protograph Embeddings (MASCHInE). In particular, MASCHInE helps produce more versatile KGEs that yield substantially better performance for entity clustering and node classification tasks. For link prediction, using MASCHInE has little impact on rank-based performance but increases the number of semantically valid predictions.
☆ Online Learning under Adversarial Nonlinear Constraints
In many applications, learning systems are required to process continuous non-stationary data streams. We study this problem in an online learning framework and propose an algorithm that can deal with adversarial time-varying and nonlinear constraints. As we show in our work, the algorithm called Constraint Violation Velocity Projection (CVV-Pro) achieves $\sqrt{T}$ regret and converges to the feasible set at a rate of $1/\sqrt{T}$, despite the fact that the feasible set is slowly time-varying and a priori unknown to the learner. CVV-Pro only relies on local sparse linear approximations of the feasible set and therefore avoids optimizing over the entire set at each iteration, which is in sharp contrast to projected gradients or Frank-Wolfe methods. We also empirically evaluate our algorithm on two-player games, where the players are subjected to a shared constraint.
☆ Supervised Knowledge May Hurt Novel Class Discovery Performance
Novel class discovery (NCD) aims to infer novel categories in an unlabeled dataset by leveraging prior knowledge of a labeled set comprising disjoint but related classes. Given that most existing literature focuses primarily on utilizing supervised knowledge from a labeled set at the methodology level, this paper considers the question: Is supervised knowledge always helpful at different levels of semantic relevance? To proceed, we first establish a novel metric, so-called transfer flow, to measure the semantic similarity between labeled/unlabeled datasets. To show the validity of the proposed metric, we build up a large-scale benchmark with various degrees of semantic similarities between labeled/unlabeled datasets on ImageNet by leveraging its hierarchical class structure. The results based on the proposed benchmark show that the proposed transfer flow is in line with the hierarchical class structure; and that NCD performance is consistent with the semantic similarities (measured by the proposed metric). Next, by using the proposed transfer flow, we conduct various empirical experiments with different levels of semantic similarity, yielding that supervised knowledge may hurt NCD performance. Specifically, using supervised information from a low-similarity labeled set may lead to a suboptimal result as compared to using pure self-supervised knowledge. These results reveal the inadequacy of the existing NCD literature which usually assumes that supervised knowledge is beneficial. Finally, we develop a pseudo-version of the transfer flow as a practical reference to decide if supervised knowledge should be used in NCD. Its effectiveness is supported by our empirical studies, which show that the pseudo transfer flow (with or without supervised knowledge) is consistent with the corresponding accuracy based on various datasets. Code is released at https://github.com/J-L-O/SK-Hurt-NCD
comment: TMLR 2023 accepted paper. arXiv admin note: substantial text overlap with arXiv:2209.09120
☆ Proximal Symmetric Non-negative Latent Factor Analysis: A Novel Approach to Highly-Accurate Representation of Undirected Weighted Networks
An Undirected Weighted Network (UWN) is commonly found in big data-related applications. Note that such a network's information connected with its nodes, and edges can be expressed as a Symmetric, High-Dimensional and Incomplete (SHDI) matrix. However, existing models fail in either modeling its intrinsic symmetry or low-data density, resulting in low model scalability or representation learning ability. For addressing this issue, a Proximal Symmetric Nonnegative Latent-factor-analysis (PSNL) model is proposed. It incorporates a proximal term into symmetry-aware and data density-oriented objective function for high representation accuracy. Then an adaptive Alternating Direction Method of Multipliers (ADMM)-based learning scheme is implemented through a Tree-structured of Parzen Estimators (TPE) method for high computational efficiency. Empirical studies on four UWNs demonstrate that PSNL achieves higher accuracy gain than state-of-the-art models, as well as highly competitive computational efficiency.
☆ Dance Generation by Sound Symbolic Words
This study introduces a novel approach to generate dance motions using onomatopoeia as input, with the aim of enhancing creativity and diversity in dance generation. Unlike text and music, onomatopoeia conveys rhythm and meaning through abstract word expressions without constraints on expression and without need for specialized knowledge. We adapt the AI Choreographer framework and employ the Sakamoto system, a feature extraction method for onomatopoeia focusing on phonemes and syllables. Additionally, we present a new dataset of 40 onomatopoeia-dance motion pairs collected through a user survey. Our results demonstrate that the proposed method enables more intuitive dance generation and can create dance motions using sound-symbolic words from a variety of languages, including those without onomatopoeia. This highlights the potential for diverse dance creation across different languages and cultures, accessible to a wider audience. Qualitative samples from our model can be found at: https://sites.google.com/view/onomatopoeia-dance/home/.
☆ Understanding Progressive Training Through the Framework of Randomized Coordinate Descent
We propose a Randomized Progressive Training algorithm (RPT) -- a stochastic proxy for the well-known Progressive Training method (PT) (Karras et al., 2017). Originally designed to train GANs (Goodfellow et al., 2014), PT was proposed as a heuristic, with no convergence analysis even for the simplest objective functions. On the contrary, to the best of our knowledge, RPT is the first PT-type algorithm with rigorous and sound theoretical guarantees for general smooth objective functions. We cast our method into the established framework of Randomized Coordinate Descent (RCD) (Nesterov, 2012; Richt\'arik & Tak\'a\v{c}, 2014), for which (as a by-product of our investigations) we also propose a novel, simple and general convergence analysis encapsulating strongly-convex, convex and nonconvex objectives. We then use this framework to establish a convergence theory for RPT. Finally, we validate the effectiveness of our method through extensive computational experiments.
☆ Fair and Robust Estimation of Heterogeneous Treatment Effects for Policy Learning
We propose a simple and general framework for nonparametric estimation of heterogeneous treatment effects under fairness constraints. Under standard regularity conditions, we show that the resulting estimators possess the double robustness property. We use this framework to characterize the trade-off between fairness and the maximum welfare achievable by the optimal policy. We evaluate the methods in a simulation study and illustrate them in a real-world case study.
☆ Spike-based computation using classical recurrent neural networks
Spiking neural networks are a type of artificial neural networks in which communication between neurons is only made of events, also called spikes. This property allows neural networks to make asynchronous and sparse computations and therefore to drastically decrease energy consumption when run on specialized hardware. However, training such networks is known to be difficult, mainly due to the non-differentiability of the spike activation, which prevents the use of classical backpropagation. This is because state-of-the-art spiking neural networks are usually derived from biologically-inspired neuron models, to which are applied machine learning methods for training. Nowadays, research about spiking neural networks focuses on the design of training algorithms whose goal is to obtain networks that compete with their non-spiking version on specific tasks. In this paper, we attempt the symmetrical approach: we modify the dynamics of a well-known, easily trainable type of recurrent neural network to make it event-based. This new RNN cell, called the Spiking Recurrent Cell, therefore communicates using events, i.e. spikes, while being completely differentiable. Vanilla backpropagation can thus be used to train any network made of such RNN cell. We show that this new network can achieve performance comparable to other types of spiking networks in the MNIST benchmark and its variants, the Fashion-MNIST and the Neuromorphic-MNIST. Moreover, we show that this new cell makes the training of deep spiking networks achievable.
comment: 12 pages, 3 figures
☆ Zero-shot Preference Learning for Offline RL via Optimal Transport
Preference-based Reinforcement Learning (PbRL) has demonstrated remarkable efficacy in aligning rewards with human intentions. However, a significant challenge lies in the need of substantial human labels, which is costly and time-consuming. Additionally, the expensive preference data obtained from prior tasks is not typically reusable for subsequent task learning, leading to extensive labeling for each new task. In this paper, we propose a novel zero-shot preference-based RL algorithm that leverages labeled preference data from source tasks to infer labels for target tasks, eliminating the requirement for human queries. Our approach utilizes Gromov-Wasserstein distance to align trajectory distributions between source and target tasks. The solved optimal transport matrix serves as a correspondence between trajectories of two tasks, making it possible to identify corresponding trajectory pairs between tasks and transfer the preference labels. However, learning directly from inferred labels that contains a fraction of noisy labels will result in an inaccurate reward function, subsequently affecting policy performance. To this end, we introduce Robust Preference Transformer, which models the rewards as Gaussian distributions and incorporates reward uncertainty in addition to reward mean. The empirical results on robotic manipulation tasks of Meta-World and Robomimic show that our method has strong capabilities of transferring preferences between tasks and learns reward functions from noisy labels robustly. Furthermore, we reveal that our method attains near-oracle performance with a small proportion of scripted labels.
☆ Buying Information for Stochastic Optimization ICML 2023
Stochastic optimization is one of the central problems in Machine Learning and Theoretical Computer Science. In the standard model, the algorithm is given a fixed distribution known in advance. In practice though, one may acquire at a cost extra information to make better decisions. In this paper, we study how to buy information for stochastic optimization and formulate this question as an online learning problem. Assuming the learner has an oracle for the original optimization problem, we design a $2$-competitive deterministic algorithm and a $e/(e-1)$-competitive randomized algorithm for buying information. We show that this ratio is tight as the problem is equivalent to a robust generalization of the ski-rental problem, which we call super-martingale stopping. We also consider an adaptive setting where the learner can choose to buy information after taking some actions for the underlying optimization problem. We focus on the classic optimization problem, Min-Sum Set Cover, where the goal is to quickly find an action that covers a given request drawn from a known distribution. We provide an $8$-competitive algorithm running in polynomial time that chooses actions and decides when to buy information about the underlying request.
comment: To appear in ICML 2023
☆ Avoid Adversarial Adaption in Federated Learning by Multi-Metric Investigations
Federated Learning (FL) trains machine learning models on data distributed across multiple devices, avoiding data transfer to a central location. This improves privacy, reduces communication costs, and enhances model performance. However, FL is prone to poisoning attacks, which can be untargeted aiming to reduce the model performance, or targeted, so-called backdoors, which add adversarial behavior that can be triggered with appropriately crafted inputs. Striving for stealthiness, backdoor attacks are harder to deal with. Mitigation techniques against poisoning attacks rely on monitoring certain metrics and filtering malicious model updates. However, previous works didn't consider real-world adversaries and data distributions. To support our statement, we define a new notion of strong adaptive adversaries that can simultaneously adapt to multiple objectives and demonstrate through extensive tests, that existing defense methods can be circumvented in this adversary model. We also demonstrate, that existing defenses have limited effectiveness when no assumptions are made about underlying data distributions. To address realistic scenarios and adversary models, we propose Metric-Cascades (MESAS) a new defense that leverages multiple detection metrics simultaneously for the filtering of poisoned model updates. This approach forces adaptive attackers into a heavy multi-objective optimization problem, and our evaluation with nine backdoors and three datasets shows that even our strong adaptive attacker cannot evade MESAS's detection. We show that MESAS outperforms existing defenses in distinguishing backdoors from distortions originating from different data distributions within and across the clients. Overall, MESAS is the first defense that is robust against strong adaptive adversaries and is effective in real-world data scenarios while introducing a low overhead of 24.37s on average.
comment: 23 pages, 12 figures, 27 tables, 11 equations
☆ How does over-squashing affect the power of GNNs?
Graph Neural Networks (GNNs) are the state-of-the-art model for machine learning on graph-structured data. The most popular class of GNNs operate by exchanging information between adjacent nodes, and are known as Message Passing Neural Networks (MPNNs). Given their widespread use, understanding the expressive power of MPNNs is a key question. However, existing results typically consider settings with uninformative node features. In this paper, we provide a rigorous analysis to determine which function classes of node features can be learned by an MPNN of a given capacity. We do so by measuring the level of pairwise interactions between nodes that MPNNs allow for. This measure provides a novel quantitative characterization of the so-called over-squashing effect, which is observed to occur when a large volume of messages is aggregated into fixed-size vectors. Using our measure, we prove that, to guarantee sufficient communication between pairs of nodes, the capacity of the MPNN must be large enough, depending on properties of the input graph structure, such as commute times. For many relevant scenarios, our analysis results in impossibility statements in practice, showing that over-squashing hinders the expressive power of MPNNs. We validate our theoretical findings through extensive controlled experiments and ablation studies.
☆ L-C2ST: Local Diagnostics for Posterior Approximations in Simulation-Based Inference
Many recent works in simulation-based inference (SBI) rely on deep generative models to approximate complex, high-dimensional posterior distributions. However, evaluating whether or not these approximations can be trusted remains a challenge. Most approaches evaluate the posterior estimator only in expectation over the observation space. This limits their interpretability and is not sufficient to identify for which observations the approximation can be trusted or should be improved. Building upon the well-known classifier two-sample test (C2ST), we introduce L-C2ST, a new method that allows for a local evaluation of the posterior estimator at any given observation. It offers theoretically grounded and easy to interpret - e.g. graphical - diagnostics, and unlike C2ST, does not require access to samples from the true posterior. In the case of normalizing flow-based posterior estimators, L-C2ST can be specialized to offer better statistical power, while being computationally more efficient. On standard SBI benchmarks, L-C2ST provides comparable results to C2ST and outperforms alternative local approaches such as coverage tests based on highest predictive density (HPD). We further highlight the importance of local evaluation and the benefit of interpretability of L-C2ST on a challenging application from computational neuroscience.
comment: 20 pages, 4 figures, 7 appendices, in proceedings
☆ Personalization Disentanglement for Federated Learning
Personalized federated learning (PFL) jointly trains a variety of local models through balancing between knowledge sharing across clients and model personalization per client. This paper addresses PFL via explicit disentangling latent representations into two parts to capture the shared knowledge and client-specific personalization, which leads to more reliable and effective PFL. The disentanglement is achieved by a novel Federated Dual Variational Autoencoder (FedDVA), which employs two encoders to infer the two types of representations. FedDVA can produce a better understanding of the trade-off between global knowledge sharing and local personalization in PFL. Moreover, it can be integrated with existing FL methods and turn them into personalized models for heterogeneous downstream tasks. Extensive experiments validate the advantages caused by disentanglement and show that models trained with disentangled representations substantially outperform those vanilla methods.
☆ Memory-Based Dual Gaussian Processes for Sequential Learning ICML
Sequential learning with Gaussian processes (GPs) is challenging when access to past data is limited, for example, in continual and active learning. In such cases, errors can accumulate over time due to inaccuracies in the posterior, hyperparameters, and inducing points, making accurate learning challenging. Here, we present a method to keep all such errors in check using the recently proposed dual sparse variational GP. Our method enables accurate inference for generic likelihoods and improves learning by actively building and updating a memory of past data. We demonstrate its effectiveness in several applications involving Bayesian optimization, active learning, and continual learning.
comment: International Conference on Machine Learning (ICML) 2023
☆ CIN++: Enhancing Topological Message Passing
Graph Neural Networks (GNNs) have demonstrated remarkable success in learning from graph-structured data. However, they face significant limitations in expressive power, struggling with long-range interactions and lacking a principled approach to modeling higher-order structures and group interactions. Cellular Isomorphism Networks (CINs) recently addressed most of these challenges with a message passing scheme based on cell complexes. Despite their advantages, CINs make use only of boundary and upper messages which do not consider a direct interaction between the rings present in the underlying complex. Accounting for these interactions might be crucial for learning representations of many real-world complex phenomena such as the dynamics of supramolecular assemblies, neural activity within the brain, and gene regulation processes. In this work, we propose CIN++, an enhancement of the topological message passing scheme introduced in CINs. Our message passing scheme accounts for the aforementioned limitations by letting the cells to receive also lower messages within each layer. By providing a more comprehensive representation of higher-order and long-range interactions, our enhanced topological message passing scheme achieves state-of-the-art results on large-scale and long-range chemistry benchmarks.
comment: 21 pages, 9 figures
☆ Machine Unlearning: A Survey
Machine learning has attracted widespread attention and evolved into an enabling technology for a wide range of highly successful applications, such as intelligent computer vision, speech recognition, medical diagnosis, and more. Yet a special need has arisen where, due to privacy, usability, and/or the right to be forgotten, information about some specific samples needs to be removed from a model, called machine unlearning. This emerging technology has drawn significant interest from both academics and industry due to its innovation and practicality. At the same time, this ambitious problem has led to numerous research efforts aimed at confronting its challenges. To the best of our knowledge, no study has analyzed this complex topic or compared the feasibility of existing unlearning solutions in different kinds of scenarios. Accordingly, with this survey, we aim to capture the key concepts of unlearning techniques. The existing solutions are classified and summarized based on their characteristics within an up-to-date and comprehensive review of each category's advantages and limitations. The survey concludes by highlighting some of the outstanding issues with unlearning techniques, along with some feasible directions for new research opportunities.
☆ State Regularized Policy Optimization on Data with Dynamics Shift
In many real-world scenarios, Reinforcement Learning (RL) algorithms are trained on data with dynamics shift, i.e., with different underlying environment dynamics. A majority of current methods address such issue by training context encoders to identify environment parameters. Data with dynamics shift are separated according to their environment parameters to train the corresponding policy. However, these methods can be sample inefficient as data are used \textit{ad hoc}, and policies trained for one dynamics cannot benefit from data collected in all other environments with different dynamics. In this paper, we find that in many environments with similar structures and different dynamics, optimal policies have similar stationary state distributions. We exploit such property and learn the stationary state distribution from data with dynamics shift for efficient data reuse. Such distribution is used to regularize the policy trained in a new environment, leading to the SRPO (\textbf{S}tate \textbf{R}egularized \textbf{P}olicy \textbf{O}ptimization) algorithm. To conduct theoretical analyses, the intuition of similar environment structures is characterized by the notion of homomorphous MDPs. We then demonstrate a lower-bound performance guarantee on policies regularized by the stationary state distribution. In practice, SRPO can be an add-on module to context-based algorithms in both online and offline RL settings. Experimental results show that SRPO can make several context-based algorithms far more data efficient and significantly improve their overall performance.
comment: Preprint. Under Review
☆ Scalable Concept Extraction in Industry 4.0
The industry 4.0 is leveraging digital technologies and machine learning techniques to connect and optimize manufacturing processes. Central to this idea is the ability to transform raw data into human understandable knowledge for reliable data-driven decision-making. Convolutional Neural Networks (CNNs) have been instrumental in processing image data, yet, their ``black box'' nature complicates the understanding of their prediction process. In this context, recent advances in the field of eXplainable Artificial Intelligence (XAI) have proposed the extraction and localization of concepts, or which visual cues intervene on the prediction process of CNNs. This paper tackles the application of concept extraction (CE) methods to industry 4.0 scenarios. To this end, we modify a recently developed technique, ``Extracting Concepts with Local Aggregated Descriptors'' (ECLAD), improving its scalability. Specifically, we propose a novel procedure for calculating concept importance, utilizing a wrapper function designed for CNNs. This process is aimed at decreasing the number of times each image needs to be evaluated. Subsequently, we demonstrate the potential of CE methods, by applying them in three industrial use cases. We selected three representative use cases in the context of quality control for material design (tailored textiles), manufacturing (carbon fiber reinforcement), and maintenance (photovoltaic module inspection). In these examples, CE was able to successfully extract and locate concepts directly related to each task. This is, the visual cues related to each concept, coincided with what human experts would use to perform the task themselves, even when the visual cues were entangled between multiple classes. Through empirical results, we show that CE can be applied for understanding CNNs in an industrial context, giving useful insights that can relate to domain knowledge.
☆ Learning Dynamical Systems from Noisy Data with Inverse-Explicit Integrators
We introduce the mean inverse integrator (MII), a novel approach to increase the accuracy when training neural networks to approximate vector fields of dynamical systems from noisy data. This method can be used to average multiple trajectories obtained by numerical integrators such as Runge-Kutta methods. We show that the class of mono-implicit Runge-Kutta methods (MIRK) has particular advantages when used in connection with MII. When training vector field approximations, explicit expressions for the loss functions are obtained when inserting the training data in the MIRK formulae, unlocking symmetric and high-order integrators that would otherwise be implicit for initial value problems. The combined approach of applying MIRK within MII yields a significantly lower error compared to the plain use of the numerical integrator without averaging the trajectories. This is demonstrated with experiments using data from several (chaotic) Hamiltonian systems. Additionally, we perform a sensitivity analysis of the loss functions under normally distributed perturbations, supporting the favorable performance of MII.
comment: 23 pages, 10 figures
☆ How to Select Which Active Learning Strategy is Best Suited for Your Specific Problem and Budget
In Active Learning (AL), a learner actively chooses which unlabeled examples to query for labels from an oracle, under some budget constraints. Different AL query strategies are more suited to different problems and budgets. Therefore, in practice, knowing in advance which AL strategy is most suited for the problem at hand remains an open problem. To tackle this challenge, we propose a practical derivative-based method that dynamically identifies the best strategy for each budget. We provide theoretical analysis of a simplified case to motivate our approach and build intuition. We then introduce a method to dynamically select an AL strategy based on the specific problem and budget. Empirical results showcase the effectiveness of our approach across diverse budgets and computer vision tasks.
☆ Masked Autoencoders are Efficient Continual Federated Learners
Machine learning is typically framed from a perspective of i.i.d., and more importantly, isolated data. In parts, federated learning lifts this assumption, as it sets out to solve the real-world challenge of collaboratively learning a shared model from data distributed across clients. However, motivated primarily by privacy and computational constraints, the fact that data may change, distributions drift, or even tasks advance individually on clients, is seldom taken into account. The field of continual learning addresses this separate challenge and first steps have recently been taken to leverage synergies in distributed supervised settings, in which several clients learn to solve changing classification tasks over time without forgetting previously seen ones. Motivated by these prior works, we posit that such federated continual learning should be grounded in unsupervised learning of representations that are shared across clients; in the loose spirit of how humans can indirectly leverage others' experience without exposure to a specific task. For this purpose, we demonstrate that masked autoencoders for distribution estimation are particularly amenable to this setup. Specifically, their masking strategy can be seamlessly integrated with task attention mechanisms to enable selective knowledge transfer between clients. We empirically corroborate the latter statement through several continual federated scenarios on both image and binary datasets.
☆ On Pitfalls of Test-Time Adaptation ICML 2023
Test-Time Adaptation (TTA) has recently emerged as a promising approach for tackling the robustness challenge under distribution shifts. However, the lack of consistent settings and systematic studies in prior literature hinders thorough assessments of existing methods. To address this issue, we present TTAB, a test-time adaptation benchmark that encompasses ten state-of-the-art algorithms, a diverse array of distribution shifts, and two evaluation protocols. Through extensive experiments, our benchmark reveals three common pitfalls in prior efforts. First, selecting appropriate hyper-parameters, especially for model selection, is exceedingly difficult due to online batch dependency. Second, the effectiveness of TTA varies greatly depending on the quality and properties of the model being adapted. Third, even under optimal algorithmic conditions, none of the existing methods are capable of addressing all common types of distribution shifts. Our findings underscore the need for future research in the field to conduct rigorous evaluations on a broader set of models and shifts, and to re-examine the assumptions behind the empirical success of TTA. Our code is available at \url{https://github.com/lins-lab/ttab}.
comment: Accepted at ICML 2023
☆ Continual Learning in Linear Classification on Separable Data
We analyze continual learning on a sequence of separable linear classification tasks with binary labels. We show theoretically that learning with weak regularization reduces to solving a sequential max-margin problem, corresponding to a special case of the Projection Onto Convex Sets (POCS) framework. We then develop upper bounds on the forgetting and other quantities of interest under various settings with recurring tasks, including cyclic and random orderings of tasks. We discuss several practical implications to popular training practices like regularization scheduling and weighting. We point out several theoretical differences between our continual classification setting and a recently studied continual regression setting.
☆ BackpropTools: A Fast, Portable Deep Reinforcement Learning Library for Continuous Control
Deep Reinforcement Learning (RL) has been demonstrated to yield capable agents and control policies in several domains but is commonly plagued by prohibitively long training times. Additionally, in the case of continuous control problems, the applicability of learned policies on real-world embedded devices is limited due to the lack of real-time guarantees and portability of existing deep learning libraries. To address these challenges, we present BackpropTools, a dependency-free, header-only, pure C++ library for deep supervised and reinforcement learning. Leveraging the template meta-programming capabilities of recent C++ standards, we provide composable components that can be tightly integrated by the compiler. Its novel architecture allows BackpropTools to be used seamlessly on a heterogeneous set of platforms, from HPC clusters over workstations and laptops to smartphones, smartwatches, and microcontrollers. Specifically, due to the tight integration of the RL algorithms with simulation environments, BackpropTools can solve popular RL problems like the Pendulum-v1 swing-up about 7 to 15 times faster in terms of wall-clock training time compared to other popular RL frameworks when using TD3. We also provide a low-overhead and parallelized interface to the MuJoCo simulator, showing that our PPO implementation achieves state of the art returns in the Ant-v4 environment while achieving a 25 to 30 percent faster wall-clock training time. Finally, we also benchmark the policy inference on a diverse set of microcontrollers and show that in most cases our optimized inference implementation is much faster than even the manufacturer's DSP libraries. To the best of our knowledge, BackpropTools enables the first-ever demonstration of training a deep RL algorithm directly on a microcontroller, giving rise to the field of Tiny Reinforcement Learning (TinyRL). Project page: https://backprop.tools
comment: Project page: https://backprop.tools
☆ Rec4Ad: A Free Lunch to Mitigate Sample Selection Bias for Ads CTR Prediction in Taobao
Click-Through Rate (CTR) prediction serves as a fundamental component in online advertising. A common practice is to train a CTR model on advertisement (ad) impressions with user feedback. Since ad impressions are purposely selected by the model itself, their distribution differs from the inference distribution and thus exhibits sample selection bias (SSB) that affects model performance. Existing studies on SSB mainly employ sample re-weighting techniques which suffer from high variance and poor model calibration. Another line of work relies on costly uniform data that is inadequate to train industrial models. Thus mitigating SSB in industrial models with a uniform-data-free framework is worth exploring. Fortunately, many platforms display mixed results of organic items (i.e., recommendations) and sponsored items (i.e., ads) to users, where impressions of ads and recommendations are selected by different systems but share the same user decision rationales. Based on the above characteristics, we propose to leverage recommendations samples as a free lunch to mitigate SSB for ads CTR model (Rec4Ad). After elaborating data augmentation, Rec4Ad learns disentangled representations with alignment and decorrelation modules for enhancement. When deployed in Taobao display advertising system, Rec4Ad achieves substantial gains in key business metrics, with a lift of up to +6.6\% CTR and +2.9\% RPM.
☆ A Functional Data Perspective and Baseline On Multi-Layer Out-of-Distribution Detection
A key feature of out-of-distribution (OOD) detection is to exploit a trained neural network by extracting statistical patterns and relationships through the multi-layer classifier to detect shifts in the expected input data distribution. Despite achieving solid results, several state-of-the-art methods rely on the penultimate or last layer outputs only, leaving behind valuable information for OOD detection. Methods that explore the multiple layers either require a special architecture or a supervised objective to do so. This work adopts an original approach based on a functional view of the network that exploits the sample's trajectories through the various layers and their statistical dependencies. It goes beyond multivariate features aggregation and introduces a baseline rooted in functional anomaly detection. In this new framework, OOD detection translates into detecting samples whose trajectories differ from the typical behavior characterized by the training set. We validate our method and empirically demonstrate its effectiveness in OOD detection compared to strong state-of-the-art baselines on computer vision benchmarks.
☆ Machine learning in and out of equilibrium
The algorithms used to train neural networks, like stochastic gradient descent (SGD), have close parallels to natural processes that navigate a high-dimensional parameter space -- for example protein folding or evolution. Our study uses a Fokker-Planck approach, adapted from statistical physics, to explore these parallels in a single, unified framework. We focus in particular on the stationary state of the system in the long-time limit, which in conventional SGD is out of equilibrium, exhibiting persistent currents in the space of network parameters. As in its physical analogues, the current is associated with an entropy production rate for any given training trajectory. The stationary distribution of these rates obeys the integral and detailed fluctuation theorems -- nonequilibrium generalizations of the second law of thermodynamics. We validate these relations in two numerical examples, a nonlinear regression network and MNIST digit classification. While the fluctuation theorems are universal, there are other aspects of the stationary state that are highly sensitive to the training details. Surprisingly, the effective loss landscape and diffusion matrix that determine the shape of the stationary distribution vary depending on the simple choice of minibatching done with or without replacement. We can take advantage of this nonequilibrium sensitivity to engineer an equilibrium stationary state for a particular application: sampling from a posterior distribution of network weights in Bayesian machine learning. We propose a new variation of stochastic gradient Langevin dynamics (SGLD) that harnesses without replacement minibatching. In an example system where the posterior is exactly known, this SGWORLD algorithm outperforms SGLD, converging to the posterior orders of magnitude faster as a function of the learning rate.
comment: 24 pages, 6 figures
☆ COPR: Consistency-Oriented Pre-Ranking for Online Advertising
Cascading architecture has been widely adopted in large-scale advertising systems to balance efficiency and effectiveness. In this architecture, the pre-ranking model is expected to be a lightweight approximation of the ranking model, which handles more candidates with strict latency requirements. Due to the gap in model capacity, the pre-ranking and ranking models usually generate inconsistent ranked results, thus hurting the overall system effectiveness. The paradigm of score alignment is proposed to regularize their raw scores to be consistent. However, it suffers from inevitable alignment errors and error amplification by bids when applied in online advertising. To this end, we introduce a consistency-oriented pre-ranking framework for online advertising, which employs a chunk-based sampling module and a plug-and-play rank alignment module to explicitly optimize consistency of ECPM-ranked results. A $\Delta NDCG$-based weighting mechanism is adopted to better distinguish the importance of inter-chunk samples in optimization. Both online and offline experiments have validated the superiority of our framework. When deployed in Taobao display advertising system, it achieves an improvement of up to +12.3\% CTR and +5.6\% RPM.
☆ Logic Diffusion for Knowledge Graph Reasoning
Most recent works focus on answering first order logical queries to explore the knowledge graph reasoning via multi-hop logic predictions. However, existing reasoning models are limited by the circumscribed logical paradigms of training samples, which leads to a weak generalization of unseen logic. To address these issues, we propose a plug-in module called Logic Diffusion (LoD) to discover unseen queries from surroundings and achieves dynamical equilibrium between different kinds of patterns. The basic idea of LoD is relation diffusion and sampling sub-logic by random walking as well as a special training mechanism called gradient adaption. Besides, LoD is accompanied by a novel loss function to further achieve the robust logical diffusion when facing noisy data in training or testing sets. Extensive experiments on four public datasets demonstrate the superiority of mainstream knowledge graph reasoning models with LoD over state-of-the-art. Moreover, our ablation study proves the general effectiveness of LoD on the noise-rich knowledge graph.
comment: 10 pages, 6 figures
☆ Subgraph Networks Based Contrastive Learning
Graph contrastive learning (GCL), as a self-supervised learning method, can solve the problem of annotated data scarcity. It mines explicit features in unannotated graphs to generate favorable graph representations for downstream tasks. Most existing GCL methods focus on the design of graph augmentation strategies and mutual information estimation operations. Graph augmentation produces augmented views by graph perturbations. These views preserve a locally similar structure and exploit explicit features. However, these methods have not considered the interaction existing in subgraphs. To explore the impact of substructure interactions on graph representations, we propose a novel framework called subgraph network-based contrastive learning (SGNCL). SGNCL applies a subgraph network generation strategy to produce augmented views. This strategy converts the original graph into an Edge-to-Node mapping network with both topological and attribute features. The single-shot augmented view is a first-order subgraph network that mines the interaction between nodes, node-edge, and edges. In addition, we also investigate the impact of the second-order subgraph augmentation on mining graph structure interactions, and further, propose a contrastive objective that fuses the first-order and second-order subgraph information. We compare SGNCL with classical and state-of-the-art graph contrastive learning methods on multiple benchmark datasets of different domains. Extensive experiments show that SGNCL achieves competitive or better performance (top three) on all datasets in unsupervised learning settings. Furthermore, SGNCL achieves the best average gain of 6.9\% in transfer learning compared to the best method. Finally, experiments also demonstrate that mining substructure interactions have positive implications for graph contrastive learning.
comment: 12 pages, 6 figures
☆ Russo-Ukrainian War: Prediction and explanation of Twitter suspension
On 24 February 2022, Russia invaded Ukraine, starting what is now known as the Russo-Ukrainian War, initiating an online discourse on social media. Twitter as one of the most popular SNs, with an open and democratic character, enables a transparent discussion among its large user base. Unfortunately, this often leads to Twitter's policy violations, propaganda, abusive actions, civil integrity violation, and consequently to user accounts' suspension and deletion. This study focuses on the Twitter suspension mechanism and the analysis of shared content and features of the user accounts that may lead to this. Toward this goal, we have obtained a dataset containing 107.7M tweets, originating from 9.8 million users, using Twitter API. We extract the categories of shared content of the suspended accounts and explain their characteristics, through the extraction of text embeddings in junction with cosine similarity clustering. Our results reveal scam campaigns taking advantage of trending topics regarding the Russia-Ukrainian conflict for Bitcoin and Ethereum fraud, spam, and advertisement campaigns. Additionally, we apply a machine learning methodology including a SHapley Additive explainability model to understand and explain how user accounts get suspended.
☆ Transition role of entangled data in quantum machine learning
Entanglement serves as the resource to empower quantum computing. Recent progress has highlighted its positive impact on learning quantum dynamics, wherein the integration of entanglement into quantum operations or measurements of quantum machine learning (QML) models leads to substantial reductions in training data size, surpassing a specified prediction error threshold. However, an analytical understanding of how the entanglement degree in data affects model performance remains elusive. In this study, we address this knowledge gap by establishing a quantum no-free-lunch (NFL) theorem for learning quantum dynamics using entangled data. Contrary to previous findings, we prove that the impact of entangled data on prediction error exhibits a dual effect, depending on the number of permitted measurements. With a sufficient number of measurements, increasing the entanglement of training data consistently reduces the prediction error or decreases the required size of the training data to achieve the same prediction error. Conversely, when few measurements are allowed, employing highly entangled data could lead to an increased prediction error. The achieved results provide critical guidance for designing advanced QML protocols, especially for those tailored for execution on early-stage quantum computers with limited access to quantum resources.
☆ GSHOT: Few-shot Generative Modeling of Labeled Graphs
Deep graph generative modeling has gained enormous attraction in recent years due to its impressive ability to directly learn the underlying hidden graph distribution. Despite their initial success, these techniques, like much of the existing deep generative methods, require a large number of training samples to learn a good model. Unfortunately, large number of training samples may not always be available in scenarios such as drug discovery for rare diseases. At the same time, recent advances in few-shot learning have opened door to applications where available training data is limited. In this work, we introduce the hitherto unexplored paradigm of few-shot graph generative modeling. Towards this, we develop GSHOT, a meta-learning based framework for few-shot labeled graph generative modeling. GSHOT learns to transfer meta-knowledge from similar auxiliary graph datasets. Utilizing these prior experiences, GSHOT quickly adapts to an unseen graph dataset through self-paced fine-tuning. Through extensive experiments on datasets from diverse domains having limited training samples, we establish that GSHOT generates graphs of superior fidelity compared to existing baselines.
☆ Convergent Bregman Plug-and-Play Image Restoration for Poisson Inverse Problems
Plug-and-Play (PnP) methods are efficient iterative algorithms for solving ill-posed image inverse problems. PnP methods are obtained by using deep Gaussian denoisers instead of the proximal operator or the gradient-descent step within proximal algorithms. Current PnP schemes rely on data-fidelity terms that have either Lipschitz gradients or closed-form proximal operators, which is not applicable to Poisson inverse problems. Based on the observation that the Gaussian noise is not the adequate noise model in this setting, we propose to generalize PnP using theBregman Proximal Gradient (BPG) method. BPG replaces the Euclidean distance with a Bregman divergence that can better capture the smoothness properties of the problem. We introduce the Bregman Score Denoiser specifically parametrized and trained for the new Bregman geometry and prove that it corresponds to the proximal operator of a nonconvex potential. We propose two PnP algorithms based on the Bregman Score Denoiser for solving Poisson inverse problems. Extending the convergence results of BPG in the nonconvex settings, we show that the proposed methods converge, targeting stationary points of an explicit global functional. Experimental evaluations conducted on various Poisson inverse problems validate the convergence results and showcase effective restoration performance.
☆ Natural Language Commanding via Program Synthesis
We present Semantic Interpreter, a natural language-friendly AI system for productivity software such as Microsoft Office that leverages large language models (LLMs) to execute user intent across application features. While LLMs are excellent at understanding user intent expressed as natural language, they are not sufficient for fulfilling application-specific user intent that requires more than text-to-text transformations. We therefore introduce the Office Domain Specific Language (ODSL), a concise, high-level language specialized for performing actions in and interacting with entities in Office applications. Semantic Interpreter leverages an Analysis-Retrieval prompt construction method with LLMs for program synthesis, translating natural language user utterances to ODSL programs that can be transpiled to application APIs and then executed. We focus our discussion primarily on a research exploration for Microsoft PowerPoint.
♻ ☆ blob loss: instance imbalance aware loss functions for semantic segmentation
Deep convolutional neural networks (CNN) have proven to be remarkably effective in semantic segmentation tasks. Most popular loss functions were introduced targeting improved volumetric scores, such as the Dice coefficient (DSC). By design, DSC can tackle class imbalance, however, it does not recognize instance imbalance within a class. As a result, a large foreground instance can dominate minor instances and still produce a satisfactory DSC. Nevertheless, detecting tiny instances is crucial for many applications, such as disease monitoring. For example, it is imperative to locate and surveil small-scale lesions in the follow-up of multiple sclerosis patients. We propose a novel family of loss functions, \emph{blob loss}, primarily aimed at maximizing instance-level detection metrics, such as F1 score and sensitivity. \emph{Blob loss} is designed for semantic segmentation problems where detecting multiple instances matters. We extensively evaluate a DSC-based \emph{blob loss} in five complex 3D semantic segmentation tasks featuring pronounced instance heterogeneity in terms of texture and morphology. Compared to soft Dice loss, we achieve 5% improvement for MS lesions, 3% improvement for liver tumor, and an average 2% improvement for microscopy segmentation tasks considering F1 score.
comment: 23 pages, 7 figures // corrected one mistake where it said beta instead of alpha in the text
♻ ☆ A Watermark for Large Language Models ICML 2023
Potential harms of large language models can be mitigated by watermarking model output, i.e., embedding signals into generated text that are invisible to humans but algorithmically detectable from a short span of tokens. We propose a watermarking framework for proprietary language models. The watermark can be embedded with negligible impact on text quality, and can be detected using an efficient open-source algorithm without access to the language model API or parameters. The watermark works by selecting a randomized set of "green" tokens before a word is generated, and then softly promoting use of green tokens during sampling. We propose a statistical test for detecting the watermark with interpretable p-values, and derive an information-theoretic framework for analyzing the sensitivity of the watermark. We test the watermark using a multi-billion parameter model from the Open Pretrained Transformer (OPT) family, and discuss robustness and security.
comment: 13 pages in the main body. Published at ICML 2023. Code is available at github.com/jwkirchenbauer/lm-watermarking
♻ ☆ Concept-based Explanations for Out-Of-Distribution Detectors ICML'23
Out-of-distribution (OOD) detection plays a crucial role in ensuring the safe deployment of deep neural network (DNN) classifiers. While a myriad of methods have focused on improving the performance of OOD detectors, a critical gap remains in interpreting their decisions. We help bridge this gap by providing explanations for OOD detectors based on learned high-level concepts. We first propose two new metrics for assessing the effectiveness of a particular set of concepts for explaining OOD detectors: 1) detection completeness, which quantifies the sufficiency of concepts for explaining an OOD-detector's decisions, and 2) concept separability, which captures the distributional separation between in-distribution and OOD data in the concept space. Based on these metrics, we propose an unsupervised framework for learning a set of concepts that satisfy the desired properties of high detection completeness and concept separability, and demonstrate its effectiveness in providing concept-based explanations for diverse off-the-shelf OOD detectors. We also show how to identify prominent concepts contributing to the detection results, and provide further reasoning about their decisions.
comment: Paper published at International Conference on Machine Learning (ICML'23)
♻ ☆ A Trustworthiness Score to Evaluate CNNs Predictions
Due to the black box nature of Convolutional Neural Networks (CNNs), the continuous validation of CNNs during operation is challenging with the absence of a human monitor. As a result this makes it difficult for developers and regulators to gain confidence in the deployment of autonomous systems employing CNNs. It is critical for safety during operation to know when CNN's predictions are trustworthy or suspicious. With the absence of a human monitor, the basic approach is to use the model's output confidence score to assess if predictions are trustworthy or suspicious. However, the model's confidence score is a result of computations coming from a black box, therefore lacks transparency and makes it challenging to automatedly credit trustworthiness to predictions. We introduce the trustworthiness score (TS), a simple metric that provides a more transparent and effective way of providing confidence in CNNs predictions compared to model's confidence score. The metric quantifies the trustworthiness in a prediction by checking for the existence of certain features in the predictions made by the CNN. We also use the underlying idea of the TS metric, to provide a suspiciousness score (SS) in the overall input frame to help in the detection of suspicious frames where false negatives exist. We conduct a case study using YOLOv5 on persons detection to demonstrate our method and usage of TS and SS. The case study shows that using our method consistently improves the precision of predictions compared to relying on model confidence score alone, for both 1) approving of trustworthy predictions (~20% improvement) and 2) detecting suspicious frames (~5% improvement).
♻ ☆ A Theory of Link Prediction via Relational Weisfeiler-Leman
Graph neural networks are prominent models for representation learning over graph-structured data. While the capabilities and limitations of these models are well-understood for simple graphs, our understanding remains incomplete in the context of knowledge graphs. Our goal is to provide a systematic understanding of the landscape of graph neural networks for knowledge graphs pertaining to the prominent task of link prediction. Our analysis entails a unifying perspective on seemingly unrelated models and unlocks a series of other models. The expressive power of various models is characterized via a corresponding relational Weisfeiler-Leman algorithm. This analysis is extended to provide a precise logical characterization of the class of functions captured by a class of graph neural networks. The theoretical findings presented in this paper explain the benefits of some widely employed practical design choices, which are validated empirically.
♻ ☆ AutoPEFT: Automatic Configuration Search for Parameter-Efficient Fine-Tuning
Large pretrained language models are widely used in downstream NLP tasks via task-specific fine-tuning, but such procedures can be costly. Recently, Parameter-Efficient Fine-Tuning (PEFT) methods have achieved strong task performance while updating a much smaller number of parameters compared to full model fine-tuning (FFT). However, it is non-trivial to make informed design choices on the PEFT configurations, such as their architecture, the number of tunable parameters, and even the layers in which the PEFT modules are inserted. Consequently, it is highly likely that the current, manually designed configurations are suboptimal in terms of their performance-efficiency trade-off. Inspired by advances in neural architecture search, we propose AutoPEFT for automatic PEFT configuration selection: we first design an expressive configuration search space with multiple representative PEFT modules as building blocks. Using multi-objective Bayesian optimisation in a low-cost setup, we then discover a Pareto-optimal set of configurations with strong performance-cost trade-offs across different numbers of parameters that are also highly transferable across different tasks. Empirically, on GLUE and SuperGLUE tasks, we show that AutoPEFT-discovered configurations significantly outperform existing PEFT methods and are on par or better than FFT, without incurring substantial training efficiency costs.
comment: 17 pages, 7 figures, 9 tables
♻ ☆ Physics Inspired Approaches To Understanding Gaussian Processes
Prior beliefs about the latent function to shape inductive biases can be incorporated into a Gaussian Process (GP) via the kernel. However, beyond kernel choices, the decision-making process of GP models remains poorly understood. In this work, we contribute an analysis of the loss landscape for GP models using methods from physics. We demonstrate $\nu$-continuity for Matern kernels and outline aspects of catastrophe theory at critical points in the loss landscape. By directly including $\nu$ in the hyperparameter optimisation for Matern kernels, we find that typical values of $\nu$ are far from optimal in terms of performance, yet prevail in the literature due to the increased computational speed. We also provide an a priori method for evaluating the effect of GP ensembles and discuss various voting approaches based on physical properties of the loss landscape. The utility of these approaches is demonstrated for various synthetic and real datasets. Our findings provide an enhanced understanding of the decision-making process behind GPs and offer practical guidance for improving their performance and interpretability in a range of applications.
comment: 9 pages, 4 figures
♻ ☆ Stochastic Gradient Descent-Induced Drift of Representation in a Two-Layer Neural Network ICML
Representational drift refers to over-time changes in neural activation accompanied by a stable task performance. Despite being observed in the brain and in artificial networks, the mechanisms of drift and its implications are not fully understood. Motivated by recent experimental findings of stimulus-dependent drift in the piriform cortex, we use theory and simulations to study this phenomenon in a two-layer linear feedforward network. Specifically, in a continual online learning scenario, we study the drift induced by the noise inherent in the Stochastic Gradient Descent (SGD). By decomposing the learning dynamics into the normal and tangent spaces of the minimum-loss manifold, we show the former corresponds to a finite variance fluctuation, while the latter could be considered as an effective diffusion process on the manifold. We analytically compute the fluctuation and the diffusion coefficients for the stimuli representations in the hidden layer as functions of network parameters and input distribution. Further, consistent with experiments, we show that the drift rate is slower for a more frequently presented stimulus. Overall, our analysis yields a theoretical framework for better understanding of the drift phenomenon in biological and artificial neural networks.
comment: International Conference on Machine Learning (ICML) 2023
♻ ☆ Overcoming Simplicity Bias in Deep Networks using a Feature Sieve ICML 2023
Simplicity bias is the concerning tendency of deep networks to over-depend on simple, weakly predictive features, to the exclusion of stronger, more complex features. This is exacerbated in real-world applications by limited training data and spurious feature-label correlations, leading to biased, incorrect predictions. We propose a direct, interventional method for addressing simplicity bias in DNNs, which we call the feature sieve. We aim to automatically identify and suppress easily-computable spurious features in lower layers of the network, thereby allowing the higher network levels to extract and utilize richer, more meaningful representations. We provide concrete evidence of this differential suppression & enhancement of relevant features on both controlled datasets and real-world images, and report substantial gains on many real-world debiasing benchmarks (11.4% relative gain on Imagenet-A; 3.2% on BAR, etc). Crucially, we do not depend on prior knowledge of spurious attributes or features, and in fact outperform many baselines that explicitly incorporate such information. We believe that our feature sieve work opens up exciting new research directions in automated adversarial feature extraction and representation learning for deep networks.
comment: Accepted at ICML 2023
♻ ☆ Functional sufficient dimension reduction through information maximization with application to classification
Considering the case where the response variable is a categorical variable and the predictor is a random function, two novel functional sufficient dimensional reduction (FSDR) methods are proposed based on mutual information and square loss mutual information. Compared to the classical FSDR methods, such as functional sliced inverse regression and functional sliced average variance estimation, the proposed methods are appealing because they are capable of estimating multiple effective dimension reduction directions in the case of a relatively small number of categories, especially for the binary response. Moreover, the proposed methods do not require the restrictive linear conditional mean assumption and the constant covariance assumption. They avoid the inverse problem of the covariance operator which is often encountered in the functional sufficient dimension reduction. The functional principal component analysis with truncation be used as a regularization mechanism. Under some mild conditions, the statistical consistency of the proposed methods is established. It is demonstrated that the two methods are competitive compared with some existing FSDR methods by simulations and real data analyses.
♻ ☆ The Power of Preconditioning in Overparameterized Low-Rank Matrix Sensing
We propose $\textsf{ScaledGD($\lambda$)}$, a preconditioned gradient descent method to tackle the low-rank matrix sensing problem when the true rank is unknown, and when the matrix is possibly ill-conditioned. Using overparametrized factor representations, $\textsf{ScaledGD($\lambda$)}$ starts from a small random initialization, and proceeds by gradient descent with a specific form of damped preconditioning to combat bad curvatures induced by overparameterization and ill-conditioning. At the expense of light computational overhead incurred by preconditioners, $\textsf{ScaledGD($\lambda$)}$ is remarkably robust to ill-conditioning compared to vanilla gradient descent ($\textsf{GD}$) even with overprameterization. Specifically, we show that, under the Gaussian design, $\textsf{ScaledGD($\lambda$)}$ converges to the true low-rank matrix at a constant linear rate after a small number of iterations that scales only logarithmically with respect to the condition number and the problem dimension. This significantly improves over the convergence rate of vanilla $\textsf{GD}$ which suffers from a polynomial dependency on the condition number. Our work provides evidence on the power of preconditioning in accelerating the convergence without hurting generalization in overparameterized learning.
♻ ☆ oBERTa: Improving Sparse Transfer Learning via improved initialization, distillation, and pruning regimes ACL 2023
In this paper, we introduce the range of oBERTa language models, an easy-to-use set of language models which allows Natural Language Processing (NLP) practitioners to obtain between 3.8 and 24.3 times faster models without expertise in model compression. Specifically, oBERTa extends existing work on pruning, knowledge distillation, and quantization and leverages frozen embeddings improves distillation and model initialization to deliver higher accuracy on a broad range of transfer tasks. In generating oBERTa, we explore how the highly optimized RoBERTa differs from the BERT for pruning during pre-training and finetuning. We find it less amenable to compression during fine-tuning. We explore the use of oBERTa on seven representative NLP tasks and find that the improved compression techniques allow a pruned oBERTa model to match the performance of BERTbase and exceed the performance of Prune OFA Large on the SQUAD V1.1 Question Answering dataset, despite being 8x and 2x, respectively faster in inference. We release our code, training regimes, and associated model for broad usage to encourage usage and experimentation
comment: SustaiNLP2023 @ ACL 2023,9 pages, 2 figures, 45 tables
♻ ☆ Discovering New Interpretable Conservation Laws as Sparse Invariants
Discovering conservation laws for a given dynamical system is important but challenging. In a theorist setup (differential equations and basis functions are both known), we propose the Sparse Invariant Detector (SID), an algorithm that auto-discovers conservation laws from differential equations. Its algorithmic simplicity allows robustness and interpretability of the discovered conserved quantities. We show that SID is able to rediscover known and even discover new conservation laws in a variety of systems. For two examples in fluid mechanics and atmospheric chemistry, SID discovers 14 and 3 conserved quantities, respectively, where only 12 and 2 were previously known to domain experts.
♻ ☆ Optimally tackling covariate shift in RKHS-based nonparametric regression
We study the covariate shift problem in the context of nonparametric regression over a reproducing kernel Hilbert space (RKHS). We focus on two natural families of covariate shift problems defined using the likelihood ratios between the source and target distributions. When the likelihood ratios are uniformly bounded, we prove that the kernel ridge regression (KRR) estimator with a carefully chosen regularization parameter is minimax rate-optimal (up to a log factor) for a large family of RKHSs with regular kernel eigenvalues. Interestingly, KRR does not require full knowledge of likelihood ratios apart from an upper bound on them. In striking contrast to the standard statistical setting without covariate shift, we also demonstrate that a naive estimator, which minimizes the empirical risk over the function class, is strictly sub-optimal under covariate shift as compared to KRR. We then address the larger class of covariate shift problems where the likelihood ratio is possibly unbounded yet has a finite second moment. Here, we propose a reweighted KRR estimator that weights samples based on a careful truncation of the likelihood ratios. Again, we are able to show that this estimator is minimax rate-optimal, up to logarithmic factors.
comment: to appear in the Annals of Statistics
♻ ☆ A Communication-efficient Algorithm with Linear Convergence for Federated Minimax Learning NeurIPS 2022
In this paper, we study a large-scale multi-agent minimax optimization problem, which models many interesting applications in statistical learning and game theory, including Generative Adversarial Networks (GANs). The overall objective is a sum of agents' private local objective functions. We first analyze an important special case, empirical minimax problem, where the overall objective approximates a true population minimax risk by statistical samples. We provide generalization bounds for learning with this objective through Rademacher complexity analysis. Then, we focus on the federated setting, where agents can perform local computation and communicate with a central server. Most existing federated minimax algorithms either require communication per iteration or lack performance guarantees with the exception of Local Stochastic Gradient Descent Ascent (SGDA), a multiple-local-update descent ascent algorithm which guarantees convergence under a diminishing stepsize. By analyzing Local SGDA under the ideal condition of no gradient noise, we show that generally it cannot guarantee exact convergence with constant stepsizes and thus suffers from slow rates of convergence. To tackle this issue, we propose FedGDA-GT, an improved Federated (Fed) Gradient Descent Ascent (GDA) method based on Gradient Tracking (GT). When local objectives are Lipschitz smooth and strongly-convex-strongly-concave, we prove that FedGDA-GT converges linearly with a constant stepsize to global $\epsilon$-approximation solution with $\mathcal{O}(\log (1/\epsilon))$ rounds of communication, which matches the time complexity of centralized GDA method. Finally, we numerically show that FedGDA-GT outperforms Local SGDA.
comment: Accepted by NeurIPS 2022
♻ ☆ Ewald-based Long-Range Message Passing for Molecular Graphs ICML 2023
Neural architectures that learn potential energy surfaces from molecular data have undergone fast improvement in recent years. A key driver of this success is the Message Passing Neural Network (MPNN) paradigm. Its favorable scaling with system size partly relies upon a spatial distance limit on messages. While this focus on locality is a useful inductive bias, it also impedes the learning of long-range interactions such as electrostatics and van der Waals forces. To address this drawback, we propose Ewald message passing: a nonlocal Fourier space scheme which limits interactions via a cutoff on frequency instead of distance, and is theoretically well-founded in the Ewald summation method. It can serve as an augmentation on top of existing MPNN architectures as it is computationally inexpensive and agnostic to architectural details. We test the approach with four baseline models and two datasets containing diverse periodic (OC20) and aperiodic structures (OE62). We observe robust improvements in energy mean absolute errors across all models and datasets, averaging 10% on OC20 and 16% on OE62. Our analysis shows an outsize impact of these improvements on structures with high long-range contributions to the ground truth energy.
comment: Published at the 40th International Conference on Machine Learning (ICML 2023)
♻ ☆ Bridging the Gap: Enhancing the Utility of Synthetic Data via Post-Processing Techniques
Acquiring and annotating suitable datasets for training deep learning models is challenging. This often results in tedious and time-consuming efforts that can hinder research progress. However, generative models have emerged as a promising solution for generating synthetic datasets that can replace or augment real-world data. Despite this, the effectiveness of synthetic data is limited by their inability to fully capture the complexity and diversity of real-world data. To address this issue, we explore the use of Generative Adversarial Networks to generate synthetic datasets for training classifiers that are subsequently evaluated on real-world images. To improve the quality and diversity of the synthetic dataset, we propose three novel post-processing techniques: Dynamic Sample Filtering, Dynamic Dataset Recycle, and Expansion Trick. In addition, we introduce a pipeline called Gap Filler (GaFi), which applies these techniques in an optimal and coordinated manner to maximise classification accuracy on real-world data. Our experiments show that GaFi effectively reduces the gap with real-accuracy scores to an error of 2.03%, 1.78%, and 3.99% on the Fashion-MNIST, CIFAR-10, and CIFAR-100 datasets, respectively. These results represent a new state of the art in Classification Accuracy Score and highlight the effectiveness of post-processing techniques in improving the quality of synthetic datasets.
♻ ☆ Prediction of Post-Operative Renal and Pulmonary Complications Using Transformers
Postoperative complications pose a significant challenge in the healthcare industry, resulting in elevated healthcare expenses and prolonged hospital stays, and in rare instances, patient mortality. To improve patient outcomes and reduce healthcare costs, healthcare providers rely on various perioperative risk scores to guide clinical decisions and prioritize care. In recent years, machine learning techniques have shown promise in predicting postoperative complications and fatality, with deep learning models achieving remarkable success in healthcare applications. However, research on the application of deep learning models to intra-operative anesthesia management data is limited. In this paper, we evaluate the performance of transformer-based models in predicting postoperative acute renal failure, postoperative pulmonary complications, and postoperative in-hospital mortality. We compare our method's performance with state-of-the-art tabular data prediction models, including gradient boosting trees and sequential attention models, on a clinical dataset. Our results demonstrate that transformer-based models can achieve superior performance in predicting postoperative complications and outperform traditional machine learning models. This work highlights the potential of deep learning techniques, specifically transformer-based models, in revolutionizing the healthcare industry's approach to postoperative care.
♻ ☆ Seeing is Believing: Brain-Inspired Modular Training for Mechanistic Interpretability
We introduce Brain-Inspired Modular Training (BIMT), a method for making neural networks more modular and interpretable. Inspired by brains, BIMT embeds neurons in a geometric space and augments the loss function with a cost proportional to the length of each neuron connection. We demonstrate that BIMT discovers useful modular neural networks for many simple tasks, revealing compositional structures in symbolic formulas, interpretable decision boundaries and features for classification, and mathematical structure in algorithmic datasets. The ability to directly see modules with the naked eye can complement current mechanistic interpretability strategies such as probes, interventions or staring at all weights.
comment: Codes are available here: https://github.com/KindXiaoming/BIMT
♻ ☆ Make Your Pre-trained Model Reversible: From Parameter to Memory Efficient Fine-Tuning
Parameter-efficient fine-tuning (PEFT) of pre-trained language models (PLMs) has emerged as a highly successful approach, with training only a small number of parameters without sacrificing performance and becoming the de-facto learning paradigm with the increasing size of PLMs. However, existing PEFT methods are not memory-efficient, because they still require caching most of the intermediate activations for the gradient calculation, akin to fine-tuning. One effective way to reduce the activation memory is to apply a reversible model, so the intermediate activations are not necessary to be cached and can be recomputed. Nevertheless, modifying a PLM to its reversible variant with PEFT is not straightforward, since the reversible model has a distinct architecture from the currently released PLMs. In this paper, we first investigate what is a key factor for the success of existing PEFT methods, and realize that it's essential to preserve the PLM's starting point when initializing a PEFT method. With this finding, we propose memory-efficient fine-tuning (MEFT) that inserts adapters into a PLM, preserving the PLM's starting point and making it reversible without additional pre-training. We evaluate MEFT on the GLUE benchmark and five question-answering tasks with various backbones, BERT, RoBERTa, BART and OPT. MEFT significantly reduces the activation memory up to 84% of full fine-tuning with a negligible amount of trainable parameters. Moreover, MEFT achieves the same score on GLUE and a comparable score on the question-answering tasks as full fine-tuning.
comment: Code at https://github.com/BaohaoLiao/mefts
♻ ☆ Revisiting Bellman Errors for Offline Model Selection ICML 2023
Offline model selection (OMS), that is, choosing the best policy from a set of many policies given only logged data, is crucial for applying offline RL in real-world settings. One idea that has been extensively explored is to select policies based on the mean squared Bellman error (MSBE) of the associated Q-functions. However, previous work has struggled to obtain adequate OMS performance with Bellman errors, leading many researchers to abandon the idea. To this end, we elucidate why previous work has seen pessimistic results with Bellman errors and identify conditions under which OMS algorithms based on Bellman errors will perform well. Moreover, we develop a new estimator of the MSBE that is more accurate than prior methods. Our estimator obtains impressive OMS performance on diverse discrete control tasks, including Atari games.
comment: Published in ICML 2023
♻ ☆ Certified Reinforcement Learning with Logic Guidance
Reinforcement Learning (RL) is a widely employed machine learning architecture that has been applied to a variety of control problems. However, applications in safety-critical domains require a systematic and formal approach to specifying requirements as tasks or goals. We propose a model-free RL algorithm that enables the use of Linear Temporal Logic (LTL) to formulate a goal for unknown continuous-state/action Markov Decision Processes (MDPs). The given LTL property is translated into a Limit-Deterministic Generalised Buchi Automaton (LDGBA), which is then used to shape a synchronous reward function on-the-fly. Under certain assumptions, the algorithm is guaranteed to synthesise a control policy whose traces satisfy the LTL specification with maximal probability.
♻ ☆ Criteria Tell You More than Ratings: Criteria Preference-Aware Light Graph Convolution for Effective Multi-Criteria Recommendation KDD
The multi-criteria (MC) recommender system, which leverages MC rating information in a wide range of e-commerce areas, is ubiquitous nowadays. Surprisingly, although graph neural networks (GNNs) have been widely applied to develop various recommender systems due to GNN's high expressive capability in learning graph representations, it has been still unexplored how to design MC recommender systems with GNNs. In light of this, we make the first attempt towards designing a GNN-aided MC recommender system. Specifically, rather than straightforwardly adopting existing GNN-based recommendation methods, we devise a novel criteria preference-aware light graph convolution CPA-LGC method, which is capable of precisely capturing the criteria preference of users as well as the collaborative signal in complex high-order connectivities. To this end, we first construct an MC expansion graph that transforms user--item MC ratings into an expanded bipartite graph to potentially learn from the collaborative signal in MC ratings. Next, to strengthen the capability of criteria preference awareness, CPA-LGC incorporates newly characterized embeddings, including user-specific criteria-preference embeddings and item-specific criterion embeddings, into our graph convolution model. Through comprehensive evaluations using four real-world datasets, we demonstrate (a) the superiority over benchmark MC recommendation methods and benchmark recommendation methods using GNNs with tremendous gains, (b) the effectiveness of core components in CPA-LGC, and (c) the computational efficiency.
comment: 12 pages, 10 figures, 5 tables; 29th ACM SIGKDD Conference on Knowledge Discovery & Data (KDD 2023) (to appear) (Please cite our conference version.)
♻ ☆ A Survey of Learning on Small Data: Generalization, Optimization, and Challenge
Learning on big data brings success for artificial intelligence (AI), but the annotation and training costs are expensive. In future, learning on small data that approximates the generalization ability of big data is one of the ultimate purposes of AI, which requires machines to recognize objectives and scenarios relying on small data as humans. A series of learning topics is going on this way such as active learning and few-shot learning. However, there are few theoretical guarantees for their generalization performance. Moreover, most of their settings are passive, that is, the label distribution is explicitly controlled by finite training resources from known distributions. This survey follows the agnostic active sampling theory under a PAC (Probably Approximately Correct) framework to analyze the generalization error and label complexity of learning on small data in model-agnostic supervised and unsupervised fashion. Considering multiple learning communities could produce small data representation and related topics have been well surveyed, we thus subjoin novel geometric representation perspectives for small data: the Euclidean and non-Euclidean (hyperbolic) mean, where the optimization solutions including the Euclidean gradients, non-Euclidean gradients, and Stein gradient are presented and discussed. Later, multiple learning communities that may be improved by learning on small data are summarized, which yield data-efficient representations, such as transfer learning, contrastive learning, graph representation learning. Meanwhile, we find that the meta-learning may provide effective parameter update policies for learning on small data. Then, we explore multiple challenging scenarios for small data, such as the weak supervision and multi-label. Finally, multiple data applications that may benefit from efficient small data representation are surveyed.
♻ ☆ Regions of Reliability in the Evaluation of Multivariate Probabilistic Forecasts ICML 2023
Multivariate probabilistic time series forecasts are commonly evaluated via proper scoring rules, i.e., functions that are minimal in expectation for the ground-truth distribution. However, this property is not sufficient to guarantee good discrimination in the non-asymptotic regime. In this paper, we provide the first systematic finite-sample study of proper scoring rules for time-series forecasting evaluation. Through a power analysis, we identify the "region of reliability" of a scoring rule, i.e., the set of practical conditions where it can be relied on to identify forecasting errors. We carry out our analysis on a comprehensive synthetic benchmark, specifically designed to test several key discrepancies between ground-truth and forecast distributions, and we gauge the generalizability of our findings to real-world tasks with an application to an electricity production problem. Our results reveal critical shortcomings in the evaluation of multivariate probabilistic forecasts as commonly performed in the literature.
comment: 47 pages, 37 figures, camera-ready version, Fortieth International Conference on Machine Learning (ICML 2023)
♻ ☆ Faster Gradient-Free Algorithms for Nonsmooth Nonconvex Stochastic Optimization ICML 2023
We consider the optimization problem of the form $\min_{x \in \mathbb{R}^d} f(x) \triangleq \mathbb{E}_{\xi} [F(x; \xi)]$, where the component $F(x;\xi)$ is $L$-mean-squared Lipschitz but possibly nonconvex and nonsmooth. The recently proposed gradient-free method requires at most $\mathcal{O}( L^4 d^{3/2} \epsilon^{-4} + \Delta L^3 d^{3/2} \delta^{-1} \epsilon^{-4})$ stochastic zeroth-order oracle complexity to find a $(\delta,\epsilon)$-Goldstein stationary point of objective function, where $\Delta = f(x_0) - \inf_{x \in \mathbb{R}^d} f(x)$ and $x_0$ is the initial point of the algorithm. This paper proposes a more efficient algorithm using stochastic recursive gradient estimators, which improves the complexity to $\mathcal{O}(L^3 d^{3/2} \epsilon^{-3}+ \Delta L^2 d^{3/2} \delta^{-1} \epsilon^{-3})$.
comment: ICML 2023
♻ ☆ Explanation-based Finetuning Makes Models More Robust to Spurious Cues
Large Language Models (LLMs) are so powerful that they sometimes learn correlations between labels and features that are irrelevant to the task, leading to poor generalization on out-of-distribution data. We propose explanation-based finetuning as a general approach to mitigate LLMs' reliance on spurious correlations. Unlike standard finetuning where the model only predicts the answer given the input, we finetune the model to additionally generate a free-text explanation supporting its answer. To evaluate our method, we finetune the model on artificially constructed training sets containing different types of spurious cues, and test it on a test set without these cues. Compared to standard finetuning, our method makes GPT-3 (davinci) remarkably more robust against spurious cues in terms of accuracy drop across four classification tasks: ComVE (+1.2), CREAK (+9.1), e-SNLI (+15.4), and SBIC (+6.5). The efficacy generalizes across multiple model families and scales, with greater gains for larger models. Finally, our method also works well with explanations generated by the model, implying its applicability to more datasets without human-written explanations.
♻ ☆ Domain Generalization for Mammographic Image Analysis via Contrastive Learning
The deep learning technique has been shown to be effective in addressing several image analysis tasks within the computer-aided diagnosis scheme for mammography. The training of an efficacious deep learning model requires large amounts of data with sufficient diversity in terms of image style and quality. In particular, the diversity of image styles may be primarily attributed to the vendor factor. However, the collection of mammograms from large and diverse vendors is very expensive and sometimes impractical. Motivatedly, a novel contrastive learning method is developed to equip the deep learning models with better generalization capability. Specifically, the multi-style and multi-view unsupervised self-learning scheme is carried out to seek robust feature embedding against various vendor styles as a pre-trained model. Afterward, the pre-trained network is further fine-tuned to the downstream tasks, e.g., mass detection, matching, BI-RADS rating, and breast density classification. The proposed method has been extensively and rigorously evaluated with mammograms from various vendor-style domains and several public datasets. The experimental results suggest that the proposed domain generalization method can effectively improve the performance of four mammographic image tasks on data from either seen or unseen domains and outperform many state-of-the-art (SOTA) generalization methods.
comment: arXiv admin note: text overlap with arXiv:2111.10827
♻ ☆ Machine Learned Calabi-Yau Metrics and Curvature
Finding Ricci-flat (Calabi-Yau) metrics is a long standing problem in geometry with deep implications for string theory and phenomenology. A new attack on this problem uses neural networks to engineer approximations to the Calabi-Yau metric within a given K\"ahler class. In this paper we investigate numerical Ricci-flat metrics over smooth and singular K3 surfaces and Calabi-Yau threefolds. Using these Ricci-flat metric approximations for the Cefal\'u family of quartic twofolds and the Dwork family of quintic threefolds, we study characteristic forms on these geometries. We observe that the numerical stability of the numerically computed topological characteristic is heavily influenced by the choice of the neural network model, in particular, we briefly discuss a different neural network model, namely Spectral networks, which correctly approximate the topological characteristic of a Calabi-Yau. Using persistent homology, we show that high curvature regions of the manifolds form clusters near the singular points. For our neural network approximations, we observe a Bogomolov--Yau type inequality $3c_2 \geq c_1^2$ and observe an identity when our geometries have isolated $A_1$ type singularities. We sketch a proof that $\chi(X~\smallsetminus~\mathrm{Sing}\,{X}) + 2~|\mathrm{Sing}\,{X}| = 24$ also holds for our numerical approximations.
comment: Version accepted for publication: 48 pages, 32 figures, 8 tables, 3 appendices
♻ ☆ The Role of Relevance in Fair Ranking SIGIR 2023
Online platforms mediate access to opportunity: relevance-based rankings create and constrain options by allocating exposure to job openings and job candidates in hiring platforms, or sellers in a marketplace. In order to do so responsibly, these socially consequential systems employ various fairness measures and interventions, many of which seek to allocate exposure based on worthiness. Because these constructs are typically not directly observable, platforms must instead resort to using proxy scores such as relevance and infer them from behavioral signals such as searcher clicks. Yet, it remains an open question whether relevance fulfills its role as such a worthiness score in high-stakes fair rankings. In this paper, we combine perspectives and tools from the social sciences, information retrieval, and fairness in machine learning to derive a set of desired criteria that relevance scores should satisfy in order to meaningfully guide fairness interventions. We then empirically show that not all of these criteria are met in a case study of relevance inferred from biased user click data. We assess the impact of these violations on the estimated system fairness and analyze whether existing fairness interventions may mitigate the identified issues. Our analyses and results surface the pressing need for new approaches to relevance collection and generation that are suitable for use in fair ranking.
comment: Published in SIGIR 2023
♻ ☆ Responsible Design Patterns for Machine Learning Pipelines
Integrating ethical practices into the AI development process for artificial intelligence (AI) is essential to ensure safe, fair, and responsible operation. AI ethics involves applying ethical principles to the entire life cycle of AI systems. This is essential to mitigate potential risks and harms associated with AI, such as algorithm biases. To achieve this goal, responsible design patterns (RDPs) are critical for Machine Learning (ML) pipelines to guarantee ethical and fair outcomes. In this paper, we propose a comprehensive framework incorporating RDPs into ML pipelines to mitigate risks and ensure the ethical development of AI systems. Our framework comprises new responsible AI design patterns for ML pipelines identified through a survey of AI ethics and data management experts and validated through real-world scenarios with expert feedback. The framework guides AI developers, data scientists, and policy-makers to implement ethical practices in AI development and deploy responsible AI systems in production.
comment: 20 pages, 4 figures, 5 tables
♻ ☆ Transforming to Yoked Neural Networks to Improve ANN Structure
Most existing classical artificial neural networks (ANN) are designed as a tree structure to imitate neural networks. In this paper, we argue that the connectivity of a tree is not sufficient to characterize a neural network. The nodes of the same level of a tree cannot be connected with each other, i.e., these neural unit cannot share information with each other, which is a major drawback of ANN. Although ANN has been significantly improved in recent years to more complex structures, such as the directed acyclic graph (DAG), these methods also have unidirectional and acyclic bias for ANN. In this paper, we propose a method to build a bidirectional complete graph for the nodes in the same level of an ANN, which yokes the nodes of the same level to formulate a neural module. We call our model as YNN in short. YNN promotes the information transfer significantly which obviously helps in improving the performance of the method. Our YNN can imitate neural networks much better compared with the traditional ANN. In this paper, we analyze the existing structural bias of ANN and propose a model YNN to efficiently eliminate such structural bias. In our model, nodes also carry out aggregation and transformation of features, and edges determine the flow of information. We further impose auxiliary sparsity constraint to the distribution of connectedness, which promotes the learned structure to focus on critical connections. Finally, based on the optimized structure, we also design small neural module structure based on the minimum cut technique to reduce the computational burden of the YNN model. This learning process is compatible with the existing networks and different tasks. The obtained quantitative experimental results reflect that the learned connectivity is superior to the traditional NN structure.
♻ ☆ Prototype-Sample Relation Distillation: Towards Replay-Free Continual Learning ICML 2023
In Continual learning (CL) balancing effective adaptation while combating catastrophic forgetting is a central challenge. Many of the recent best-performing methods utilize various forms of prior task data, e.g. a replay buffer, to tackle the catastrophic forgetting problem. Having access to previous task data can be restrictive in many real-world scenarios, for example when task data is sensitive or proprietary. To overcome the necessity of using previous tasks' data, in this work, we start with strong representation learning methods that have been shown to be less prone to forgetting. We propose a holistic approach to jointly learn the representation and class prototypes while maintaining the relevance of old class prototypes and their embedded similarities. Specifically, samples are mapped to an embedding space where the representations are learned using a supervised contrastive loss. Class prototypes are evolved continually in the same latent space, enabling learning and prediction at any point. To continually adapt the prototypes without keeping any prior task data, we propose a novel distillation loss that constrains class prototypes to maintain relative similarities as compared to new task data. This method yields state-of-the-art performance in the task-incremental setting, outperforming methods relying on large amounts of data, and provides strong performance in the class-incremental setting without using any stored data points.
comment: Accepted at ICML 2023
♻ ☆ Toward Efficient Gradient-Based Value Estimation
Gradient-based methods for value estimation in reinforcement learning have favorable stability properties, but they are typically much slower than Temporal Difference (TD) learning methods. We study the root causes of this slowness and show that Mean Square Bellman Error (MSBE) is an ill-conditioned loss function in the sense that its Hessian has large condition-number. To resolve the adverse effect of poor conditioning of MSBE on gradient based methods, we propose a low complexity batch-free proximal method that approximately follows the Gauss-Newton direction and is asymptotically robust to parameterization. Our main algorithm, called RANS, is efficient in the sense that it is significantly faster than the residual gradient methods while having almost the same computational complexity, and is competitive with TD on the classic problems that we tested.
♻ ☆ UnRectDepthNet: Self-Supervised Monocular Depth Estimation using a Generic Framework for Handling Common Camera Distortion Models IROS 2020
In classical computer vision, rectification is an integral part of multi-view depth estimation. It typically includes epipolar rectification and lens distortion correction. This process simplifies the depth estimation significantly, and thus it has been adopted in CNN approaches. However, rectification has several side effects, including a reduced field of view (FOV), resampling distortion, and sensitivity to calibration errors. The effects are particularly pronounced in case of significant distortion (e.g., wide-angle fisheye cameras). In this paper, we propose a generic scale-aware self-supervised pipeline for estimating depth, euclidean distance, and visual odometry from unrectified monocular videos. We demonstrate a similar level of precision on the unrectified KITTI dataset with barrel distortion comparable to the rectified KITTI dataset. The intuition being that the rectification step can be implicitly absorbed within the CNN model, which learns the distortion model without increasing complexity. Our approach does not suffer from a reduced field of view and avoids computational costs for rectification at inference time. To further illustrate the general applicability of the proposed framework, we apply it to wide-angle fisheye cameras with 190$^\circ$ horizontal field of view. The training framework UnRectDepthNet takes in the camera distortion model as an argument and adapts projection and unprojection functions accordingly. The proposed algorithm is evaluated further on the KITTI rectified dataset, and we achieve state-of-the-art results that improve upon our previous work FisheyeDistanceNet. Qualitative results on a distorted test scene video sequence indicate excellent performance https://youtu.be/K6pbx3bU4Ss.
comment: Minor fixes added after IROS 2020 Camera ready submission. IROS 2020 presentation video - https://www.youtube.com/watch?v=3Br2KSWZRrY
♻ ☆ A Generalized Alternating Method for Bilevel Learning under the Polyak-Łojasiewicz Condition
Bilevel optimization has recently regained interest owing to its applications in emerging machine learning fields such as hyperparameter optimization, meta-learning, and reinforcement learning. Recent results have shown that simple alternating (implicit) gradient-based algorithms can achieve the same convergence rate of single-level gradient descent (GD) for bilevel problems with a strongly convex lower-level objective. However, it remains unclear whether this result can be generalized to bilevel problems beyond this basic setting. In this paper, we propose a Generalized ALternating mEthod for bilevel opTimization (GALET) with a nonconvex lower-level objective that satisfies the Polyak-{\L}ojasiewicz (PL) condition. We first introduce a stationary metric for the considered bilevel problems, which generalizes the existing metric. We then establish that GALET achieves an $\epsilon$-stationary metric for the considered problem within $\tilde{\cal O}(\epsilon^{-1})$ iterations, which matches the iteration complexity of GD for smooth nonconvex problems.
♻ ☆ Hiding in Plain Sight: Disguising Data Stealing Attacks in Federated Learning
Malicious server (MS) attacks have enabled the scaling of data stealing in federated learning to large batch sizes and secure aggregation, settings previously considered private. However, many concerns regarding client-side detectability of MS attacks were raised, questioning their practicality once they are publicly known. In this work, for the first time, we thoroughly study the problem of client-side detectability.We demonstrate that most prior MS attacks, which fundamentally rely on one of two key principles, are detectable by principled client-side checks. Further, we formulate desiderata for practical MS attacks and propose SEER, a novel attack framework that satisfies all desiderata, while stealing user data from gradients of realistic networks, even for large batch sizes (up to 512 in our experiments) and under secure aggregation. The key insight of SEER is the use of a secret decoder, which is jointly trained with the shared model. Our work represents a promising first step towards more principled treatment of MS attacks, paving the way for realistic data stealing that can compromise user privacy in real-world deployments.
♻ ☆ No-Regret Caching via Online Mirror Descent
We study an online caching problem in which requests can be served by a local cache to avoid retrieval costs from a remote server. The cache can update its state after a batch of requests and store an arbitrarily small fraction of each file. We study no-regret algorithms based on Online Mirror Descent (OMD) strategies. We show that bounds for the regret crucially depend on the diversity of the request process, provided by the diversity ratio R/h, where R is the size of the batch, and h is the maximum multiplicity of a request in a given batch. We characterize the optimality of OMD caching policies w.r.t. regret under different diversity regimes. We also prove that, when the cache must store the entire file, rather than a fraction, OMD strategies can be coupled with a randomized rounding scheme that preserves regret guarantees, even when update costs cannot be neglected. We provide a formal characterization of the rounding problem through optimal transport theory, and moreover we propose a computationally efficient randomized rounding scheme.
♻ ☆ MERT: Acoustic Music Understanding Model with Large-Scale Self-supervised Training
Self-supervised learning (SSL) has recently emerged as a promising paradigm for training generalisable models on large-scale data in the fields of vision, text, and speech. Although SSL has been proven effective in speech and audio, its application to music audio has yet to be thoroughly explored. This is primarily due to the distinctive challenges associated with modelling musical knowledge, particularly its tonal and pitched characteristics of music. To address this research gap, we propose an acoustic Music undERstanding model with large-scale self-supervised Training (MERT), which incorporates teacher models to provide pseudo labels in the masked language modelling (MLM) style acoustic pre-training. In our exploration, we identified a superior combination of teacher models, which outperforms conventional speech and audio approaches in terms of performance. This combination includes an acoustic teacher based on Residual Vector Quantization - Variational AutoEncoder (RVQ-VAE) and a musical teacher based on the Constant-Q Transform (CQT). These teachers effectively guide our student model, a BERT-style transformer encoder, to better model music audio. In addition, we introduce an in-batch noise mixture augmentation to enhance the representation robustness. Furthermore, we explore a wide range of settings to overcome the instability in acoustic language model pre-training, which allows our designed paradigm to scale from 95M to 330M parameters. Experimental results indicate that our model can generalise and perform well on 14 music understanding tasks and attains state-of-the-art (SOTA) overall scores. The code and models are online: https://github.com/yizhilll/MERT.
♻ ☆ ConCerNet: A Contrastive Learning Based Framework for Automated Conservation Law Discovery and Trustworthy Dynamical System Prediction ICML 2023
Deep neural networks (DNN) have shown great capacity of modeling a dynamical system; nevertheless, they usually do not obey physics constraints such as conservation laws. This paper proposes a new learning framework named ConCerNet to improve the trustworthiness of the DNN based dynamics modeling to endow the invariant properties. ConCerNet consists of two steps: (i) a contrastive learning method to automatically capture the system invariants (i.e. conservation properties) along the trajectory observations; (ii) a neural projection layer to guarantee that the learned dynamics models preserve the learned invariants. We theoretically prove the functional relationship between the learned latent representation and the unknown system invariant function. Experiments show that our method consistently outperforms the baseline neural networks in both coordinate error and conservation metrics by a large margin. With neural network based parameterization and no dependence on prior knowledge, our method can be extended to complex and large-scale dynamics by leveraging an autoencoder.
comment: Accepted by ICML 2023
♻ ☆ Tier Balancing: Towards Dynamic Fairness over Underlying Causal Factors
The pursuit of long-term fairness involves the interplay between decision-making and the underlying data generating process. In this paper, through causal modeling with a directed acyclic graph (DAG) on the decision-distribution interplay, we investigate the possibility of achieving long-term fairness from a dynamic perspective. We propose Tier Balancing, a technically more challenging but more natural notion to achieve in the context of long-term, dynamic fairness analysis. Different from previous fairness notions that are defined purely on observed variables, our notion goes one step further, capturing behind-the-scenes situation changes on the unobserved latent causal factors that directly carry out the influence from the current decision to the future data distribution. Under the specified dynamics, we prove that in general one cannot achieve the long-term fairness goal only through one-step interventions. Furthermore, in the effort of approaching long-term fairness, we consider the mission of "getting closer to" the long-term fairness goal and present possibility and impossibility results accordingly.
♻ ☆ Block-wise Training of Residual Networks via the Minimizing Movement Scheme AAAI 2022
End-to-end backpropagation has a few shortcomings: it requires loading the entire model during training, which can be impossible in constrained settings, and suffers from three locking problems (forward locking, update locking and backward locking), which prohibit training the layers in parallel. Solving layer-wise optimization problems can address these problems and has been used in on-device training of neural networks. We develop a layer-wise training method, particularly welladapted to ResNets, inspired by the minimizing movement scheme for gradient flows in distribution space. The method amounts to a kinetic energy regularization of each block that makes the blocks optimal transport maps and endows them with regularity. It works by alleviating the stagnation problem observed in layer-wise training, whereby greedily-trained early layers overfit and deeper layers stop increasing test accuracy after a certain depth. We show on classification tasks that the test accuracy of block-wise trained ResNets is improved when using our method, whether the blocks are trained sequentially or in parallel.
comment: 1st International Workshop on Practical Deep Learning in the Wild at AAAI 2022
♻ ☆ Provable Dynamic Fusion for Low-Quality Multimodal Data ICML 2023
The inherent challenge of multimodal fusion is to precisely capture the cross-modal correlation and flexibly conduct cross-modal interaction. To fully release the value of each modality and mitigate the influence of low-quality multimodal data, dynamic multimodal fusion emerges as a promising learning paradigm. Despite its widespread use, theoretical justifications in this field are still notably lacking. Can we design a provably robust multimodal fusion method? This paper provides theoretical understandings to answer this question under a most popular multimodal fusion framework from the generalization perspective. We proceed to reveal that several uncertainty estimation solutions are naturally available to achieve robust multimodal fusion. Then a novel multimodal fusion framework termed Quality-aware Multimodal Fusion (QMF) is proposed, which can improve the performance in terms of classification accuracy and model robustness. Extensive experimental results on multiple benchmarks can support our findings.
comment: Accepted by ICML 2023
♻ ☆ Learning to predict 3D rotational dynamics from images of a rigid body with unknown mass distribution
In many real-world settings, image observations of freely rotating 3D rigid bodies, may be available when low-dimensional measurements are not. However, the high-dimensionality of image data precludes the use of classical estimation techniques to learn the dynamics. The usefulness of standard deep learning methods is also limited because an image of a rigid body reveals nothing about the distribution of mass inside the body, which, together with initial angular velocity, is what determines how the body will rotate. We present a physics-informed neural network model to estimate and predict 3D rotational dynamics from image sequences. We achieve this using a multi-stage prediction pipeline that maps individual images to a latent representation homeomorphic to $\mathbf{SO}(3)$, computes angular velocities from latent pairs, and predicts future latent states using the Hamiltonian equations of motion. We demonstrate the efficacy of our approach on new rotating rigid-body datasets of sequences of synthetic images of rotating objects, including cubes, prisms and satellites, with unknown uniform and non-uniform mass distributions.
comment: 21 pages, 9 figures
♻ ☆ Fast Rates for Maximum Entropy Exploration ICML-2023
We address the challenge of exploration in reinforcement learning (RL) when the agent operates in an unknown environment with sparse or no rewards. In this work, we study the maximum entropy exploration problem of two different types. The first type is visitation entropy maximization previously considered by Hazan et al.(2019) in the discounted setting. For this type of exploration, we propose a game-theoretic algorithm that has $\widetilde{\mathcal{O}}(H^3S^2A/\varepsilon^2)$ sample complexity thus improving the $\varepsilon$-dependence upon existing results, where $S$ is a number of states, $A$ is a number of actions, $H$ is an episode length, and $\varepsilon$ is a desired accuracy. The second type of entropy we study is the trajectory entropy. This objective function is closely related to the entropy-regularized MDPs, and we propose a simple algorithm that has a sample complexity of order $\widetilde{\mathcal{O}}(\mathrm{poly}(S,A,H)/\varepsilon)$. Interestingly, it is the first theoretical result in RL literature that establishes the potential statistical advantage of regularized MDPs for exploration. Finally, we apply developed regularization techniques to reduce sample complexity of visitation entropy maximization to $\widetilde{\mathcal{O}}(H^2SA/\varepsilon^2)$, yielding a statistical separation between maximum entropy exploration and reward-free exploration.
comment: ICML-2023
♻ ☆ From Random Search to Bandit Learning in Metric Measure Spaces
Random Search is one of the most widely-used method for Hyperparameter Optimization, and is critical to the success of deep learning models. Despite its astonishing performance, little non-heuristic theory has been developed to describe the underlying working mechanism. This paper gives a theoretical accounting of Random Search. We introduce the concept of \emph{scattering dimension} that describes the landscape of the underlying function, and quantifies the performance of random search. We show that, when the environment is noise-free, the output of random search converges to the optimal value in probability at rate $ \widetilde{\mathcal{O}} \left( \left( \frac{1}{T} \right)^{ \frac{1}{d_s} } \right) $, where $ d_s \ge 0 $ is the scattering dimension of the underlying function. When the observed function values are corrupted by bounded $iid$ noise, the output of random search converges to the optimal value in probability at rate $ \widetilde{\mathcal{O}} \left( \left( \frac{1}{T} \right)^{ \frac{1}{d_s + 1} } \right) $. In addition, based on the principles of random search, we introduce an algorithm, called BLiN-MOS, for Lipschitz bandits in doubling metric spaces that are also endowed with a Borel measure, and show that BLiN-MOS achieves a regret rate of order $ \widetilde{\mathcal{O}} \left( T^{ \frac{d_z}{d_z + 1} } \right) $, where $d_z$ is the zooming dimension of the problem instance. Our results show that under certain conditions, the known information-theoretical lower bounds for Lipschitz bandits $\Omega \left( T^{\frac{d_z+1}{d_z+2}} \right)$ can be improved.
♻ ☆ Fisher Information Embedding for Node and Graph Learning ICML 2023
Attention-based graph neural networks (GNNs), such as graph attention networks (GATs), have become popular neural architectures for processing graph-structured data and learning node embeddings. Despite their empirical success, these models rely on labeled data and the theoretical properties of these models have yet to be fully understood. In this work, we propose a novel attention-based node embedding framework for graphs. Our framework builds upon a hierarchical kernel for multisets of subgraphs around nodes (e.g. neighborhoods) and each kernel leverages the geometry of a smooth statistical manifold to compare pairs of multisets, by "projecting" the multisets onto the manifold. By explicitly computing node embeddings with a manifold of Gaussian mixtures, our method leads to a new attention mechanism for neighborhood aggregation. We provide theoretical insights into generalizability and expressivity of our embeddings, contributing to a deeper understanding of attention-based GNNs. We propose both efficient unsupervised and supervised methods for learning the embeddings. Through experiments on several node classification benchmarks, we demonstrate that our proposed method outperforms existing attention-based graph models like GATs. Our code is available at https://github.com/BorgwardtLab/fisher_information_embedding.
comment: ICML 2023
♻ ☆ Learning Gaussian Mixture Representations for Tensor Time Series Forecasting IJCAI 2023
Tensor time series (TTS) data, a generalization of one-dimensional time series on a high-dimensional space, is ubiquitous in real-world scenarios, especially in monitoring systems involving multi-source spatio-temporal data (e.g., transportation demands and air pollutants). Compared to modeling time series or multivariate time series, which has received much attention and achieved tremendous progress in recent years, tensor time series has been paid less effort. Properly coping with the tensor time series is a much more challenging task, due to its high-dimensional and complex inner structure. In this paper, we develop a novel TTS forecasting framework, which seeks to individually model each heterogeneity component implied in the time, the location, and the source variables. We name this framework as GMRL, short for Gaussian Mixture Representation Learning. Experiment results on two real-world TTS datasets verify the superiority of our approach compared with the state-of-the-art baselines. Code and data are published on https://github.com/beginner-sketch/GMRL.
comment: Accepted by IJCAI 2023 Main Track
♻ ☆ Aligning Language Models with Preferences through f-divergence Minimization
Aligning language models with preferences can be posed as approximating a target distribution representing some desired behavior. Existing approaches differ both in the functional form of the target distribution and the algorithm used to approximate it. For instance, Reinforcement Learning from Human Feedback (RLHF) corresponds to minimizing a reverse KL from an implicit target distribution arising from a KL penalty in the objective. On the other hand, Generative Distributional Control (GDC) has an explicit target distribution and minimizes a forward KL from it using the Distributional Policy Gradient (DPG) algorithm. In this paper, we propose a new approach, f-DPG, which allows the use of any f-divergence to approximate any target distribution that can be evaluated. f-DPG unifies both frameworks (RLHF, GDC) and the approximation methods (DPG, RL with KL penalties). We show the practical benefits of various choices of divergence objectives and demonstrate that there is no universally optimal objective but that different divergences present different alignment and diversity trade-offs. We show that Jensen-Shannon divergence strikes a good balance between these objectives, and frequently outperforms forward KL divergence by a wide margin, leading to significant improvements over prior work. These distinguishing characteristics between divergences persist as the model size increases, highlighting the importance of selecting appropriate divergence objectives.
♻ ☆ Homomorphism Autoencoder -- Learning Group Structured Representations from Observed Transitions ICML2023
How can agents learn internal models that veridically represent interactions with the real world is a largely open question. As machine learning is moving towards representations containing not just observational but also interventional knowledge, we study this problem using tools from representation learning and group theory. We propose methods enabling an agent acting upon the world to learn internal representations of sensory information that are consistent with actions that modify it. We use an autoencoder equipped with a group representation acting on its latent space, trained using an equivariance-derived loss in order to enforce a suitable homomorphism property on the group representation. In contrast to existing work, our approach does not require prior knowledge of the group and does not restrict the set of actions the agent can perform. We motivate our method theoretically, and show empirically that it can learn a group representation of the actions, thereby capturing the structure of the set of transformations applied to the environment. We further show that this allows agents to predict the effect of sequences of future actions with improved accuracy.
comment: ICML2023, 26 pages, 17 figures
♻ ☆ Distribution-Free Matrix Prediction Under Arbitrary Missing Pattern
This paper studies the open problem of conformalized entry prediction in a row/column-exchangeable matrix. The matrix setting presents novel and unique challenges, but there exists little work on this interesting topic. We meticulously define the problem, differentiate it from closely related problems, and rigorously delineate the boundary between achievable and impossible goals. We then propose two practical algorithms. The first method provides a fast emulation of the full conformal prediction, while the second method leverages the technique of algorithmic stability for acceleration. Both methods are computationally efficient and can effectively safeguard coverage validity in presence of arbitrary missing pattern. Further, we quantify the impact of missingness on prediction accuracy and establish fundamental limit results. Empirical evidence from synthetic and real-world data sets corroborates the superior performance of our proposed methods.
comment: 12 pages, 4 figures
♻ ☆ A Unification Framework for Euclidean and Hyperbolic Graph Neural Networks
Hyperbolic neural networks can effectively capture the inherent hierarchy of graph datasets, and consequently a powerful choice of GNNs. However, they entangle multiple incongruent (gyro-)vector spaces within a layer, which makes them limited in terms of generalization and scalability. In this work, we propose the Poincare disk model as our search space, and apply all approximations on the disk (as if the disk is a tangent space derived from the origin), thus getting rid of all inter-space transformations. Such an approach enables us to propose a hyperbolic normalization layer and to further simplify the entire hyperbolic model to a Euclidean model cascaded with our hyperbolic normalization layer. We applied our proposed nonlinear hyperbolic normalization to the current state-of-the-art homogeneous and multi-relational graph networks. We demonstrate that our model not only leverages the power of Euclidean networks such as interpretability and efficient execution of various model components, but also outperforms both Euclidean and hyperbolic counterparts on various benchmarks. Our code is made publicly available at https://github.com/oom-debugger/ijcai23.
♻ ☆ An Evidential Real-Time Multi-Mode Fault Diagnosis Approach Based on Broad Learning System
Fault diagnosis is a crucial area of research in industry. Industrial processes exhibit diverse operating conditions, where data often have non-Gaussian, multi-mode, and center-drift characteristics. Data-driven approaches are currently the main focus in the field, but continuous fault classification and parameter updates of fault classifiers pose challenges for multiple operating modes and real-time settings. Thus, a pressing issue is to achieve real-time multi-mode fault diagnosis in industrial systems. In this paper, a novel approach to achieve real-time multi-mode fault diagnosis is proposed for industrial applications, which addresses this critical research problem. Our approach uses an extended evidence reasoning (ER) algorithm to fuse information and merge outputs from different base classifiers. These base classifiers based on broad learning system (BLS) are trained to ensure maximum fault diagnosis accuracy. Furthermore, pseudo-label learning is used to update model parameters in real-time. The effectiveness of the proposed approach is demonstrated on the multi-mode Tennessee Eastman process dataset.
comment: 6 pages, 11 figures, Accepted by the 34th Chinese Process Control Conference
♻ ☆ Rigid body flows for sampling molecular crystal structures
Normalizing flows (NF) are a class of powerful generative models that have gained popularity in recent years due to their ability to model complex distributions with high flexibility and expressiveness. In this work, we introduce a new type of normalizing flow that is tailored for modeling positions and orientations of multiple objects in three-dimensional space, such as molecules in a crystal. Our approach is based on two key ideas: first, we define smooth and expressive flows on the group of unit quaternions, which allows us to capture the continuous rotational motion of rigid bodies; second, we use the double cover property of unit quaternions to define a proper density on the rotation group. This ensures that our model can be trained using standard likelihood-based methods or variational inference with respect to a thermodynamic target density. We evaluate the method by training Boltzmann generators for two molecular examples, namely the multi-modal density of a tetrahedral system in an external field and the ice XI phase in the TIP4P water model. Our flows can be combined with flows operating on the internal degrees of freedom of molecules, and constitute an important step towards the modeling of distributions of many interacting molecules.
♻ ☆ Can In-context Learners Learn a Reasoning Concept from Demonstrations? ACL 2023
Large language models show an emergent ability to learn a new task from a small number of input-output demonstrations. However, recent work shows that in-context learners largely rely on their pre-trained knowledge, such as the sentiment of the labels, instead of finding new associations in the input. However, the commonly-used few-shot evaluation settings using a random selection of in-context demonstrations can not disentangle models' ability to learn a new skill from demonstrations, as most of the randomly-selected demonstrations do not present relations informative for prediction beyond exposing the new task distribution. To disentangle models' in-context learning ability independent of models' memory, we introduce a Conceptual few-shot learning method selecting the demonstrations sharing a possibly-informative concept with the predicted sample. We extract a set of such concepts from annotated explanations and measure how much can models benefit from presenting these concepts in few-shot demonstrations. We find that smaller models are more sensitive to the presented concepts. While some of the models are able to benefit from concept-presenting demonstrations for each assessed concept, we find that none of the assessed in-context learners can benefit from all presented reasoning concepts consistently, leaving the in-context concept learning an open challenge.
comment: Accepted at ACL 2023 Natural Language Reasoning workshop
♻ ☆ Predicting malaria dynamics in Burundi using deep Learning Models
Malaria continues to be a major public health problem on the African continent, particularly in Sub-Saharan Africa. Nonetheless, efforts are ongoing, and significant progress has been made. In Burundi, malaria is among the main public health concerns. In the literature, there are limited prediction models for Burundi. We know that such tools are much needed for interventions design. In our study, we built machine-learning based models to estimates malaria cases in Burundi. The forecast of malaria cases was carried out at province level and national scale as well. Long short term memory (LSTM) model, a type of deep learning model has been used to achieve best results using climate-change related factors such as temperature, rainfal, and relative humidity, together with malaria historical data and human population. With this model, the results showed that at country level different tuning of parameters can be used in order to determine the minimum and maximum expected malaria cases. The univariate version of that model (LSTM) which learns from previous dynamics of malaria cases give more precise estimates at province-level, but both models have same trends overall at provnce-level and country-level
♻ ☆ Safe AI for health and beyond -- Monitoring to transform a health service
Machine learning techniques are effective for building predictive models because they identify patterns in large datasets. Development of a model for complex real-life problems often stop at the point of publication, proof of concept or when made accessible through some mode of deployment. However, a model in the medical domain risks becoming obsolete as patient demographics, systems and clinical practices change. The maintenance and monitoring of predictive model performance post-publication is crucial to enable their safe and effective long-term use. We will assess the infrastructure required to monitor the outputs of a machine learning algorithm, and present two scenarios with examples of monitoring and updates of models, firstly on a breast cancer prognosis model trained on public longitudinal data, and secondly on a neurodegenerative stratification algorithm that is currently being developed and tested in clinic.
comment: 12 pages, 3 figures
♻ ☆ Graph2topic: an opensource topic modeling framework based on sentence embedding and community detection
It has been reported that clustering-based topic models, which cluster high-quality sentence embeddings with an appropriate word selection method, can generate better topics than generative probabilistic topic models. However, these approaches suffer from the inability to select appropriate parameters and incomplete models that overlook the quantitative relation between words with topics and topics with text. To solve these issues, we propose graph to topic (G2T), a simple but effective framework for topic modelling. The framework is composed of four modules. First, document representation is acquired using pretrained language models. Second, a semantic graph is constructed according to the similarity between document representations. Third, communities in document semantic graphs are identified, and the relationship between topics and documents is quantified accordingly. Fourth, the word--topic distribution is computed based on a variant of TFIDF. Automatic evaluation suggests that G2T achieved state-of-the-art performance on both English and Chinese documents with different lengths.
comment: 11pages
♻ ☆ Diffusion Models and Semi-Supervised Learners Benefit Mutually with Few Labels
In an effort to further advance semi-supervised generative and classification tasks, we propose a simple yet effective training strategy called dual pseudo training (DPT), built upon strong semi-supervised learners and diffusion models. DPT operates in three stages: training a classifier on partially labeled data to predict pseudo-labels; training a conditional generative model using these pseudo-labels to generate pseudo images; and retraining the classifier with a mix of real and pseudo images. Empirically, DPT consistently achieves SOTA performance of semi-supervised generation and classification across various settings. In particular, with one or two labels per class, DPT achieves a Fr\'echet Inception Distance (FID) score of 3.08 or 2.52 on ImageNet 256x256, surpassing strong diffusion models with full labels, such as IDDPM, CDM, ADM, and LDM. Besides, DPT outperforms competitive semi-supervised baselines substantially on ImageNet classification tasks, achieving top-1 accuracies of 59.0 (+2.8), 69.5 (+3.0), and 74.4 (+2.0) with one, two, or five labels per class, respectively. Notably, our results demonstrate that diffusion can generate realistic images with only a few labels (e.g., <0.1%) and generative augmentation remains viable for semi-supervised classification.
♻ ☆ Topological Data Analysis for Speech Processing INTERSPEECH 2023
We apply topological data analysis (TDA) to speech classification problems and to the introspection of a pretrained speech model, HuBERT. To this end, we introduce a number of topological and algebraic features derived from Transformer attention maps and embeddings. We show that a simple linear classifier built on top of such features outperforms a fine-tuned classification head. In particular, we achieve an improvement of about $9\%$ accuracy and $5\%$ ERR on four common datasets; on CREMA-D, the proposed feature set reaches a new state of the art performance with accuracy $80.155$. We also show that topological features are able to reveal functional roles of speech Transformer heads; e.g., we find the heads capable to distinguish between pairs of sample sources (natural/synthetic) or voices without any downstream fine-tuning. Our results demonstrate that TDA is a promising new approach for speech analysis, especially for tasks that require structural prediction. Appendices, an introduction to TDA, and other additional materials are available here - https://topohubert.github.io/speech-topology-webpages/
comment: Accepted to INTERSPEECH 2023 conference
♻ ☆ SELTO: Sample-Efficient Learned Topology Optimization
Recent developments in Deep Learning (DL) suggest a vast potential for Topology Optimization (TO). However, while there are some promising attempts, the subfield still lacks a firm footing regarding basic methods and datasets. We aim to address both points. First, we explore physics-based preprocessing and equivariant networks to create sample-efficient components for TO DL pipelines. We evaluate them in a large-scale ablation study using end-to-end supervised training. The results demonstrate a drastic improvement in sample efficiency and the predictions' physical correctness. Second, to improve comparability and future progress, we publish the two first TO datasets containing problems and corresponding ground truth solutions.
comment: 25 pages, 10 figures, submitted to the International Journal for Numerical Methods in Engineering
♻ ☆ Communication-Constrained Bandits under Additive Gaussian Noise
We study a distributed stochastic multi-armed bandit where a client supplies the learner with communication-constrained feedback based on the rewards for the corresponding arm pulls. In our setup, the client must encode the rewards such that the second moment of the encoded rewards is no more than $P$, and this encoded reward is further corrupted by additive Gaussian noise of variance $\sigma^2$; the learner only has access to this corrupted reward. For this setting, we derive an information-theoretic lower bound of $\Omega\left(\sqrt{\frac{KT}{\mathtt{SNR} \wedge1}} \right)$ on the minimax regret of any scheme, where $ \mathtt{SNR} := \frac{P}{\sigma^2}$, and $K$ and $T$ are the number of arms and time horizon, respectively. Furthermore, we propose a multi-phase bandit algorithm, $\mathtt{UE\text{-}UCB++}$, which matches this lower bound to a minor additive factor. $\mathtt{UE\text{-}UCB++}$ performs uniform exploration in its initial phases and then utilizes the {\em upper confidence bound }(UCB) bandit algorithm in its final phase. An interesting feature of $\mathtt{UE\text{-}UCB++}$ is that the coarser estimates of the mean rewards formed during a uniform exploration phase help to refine the encoding protocol in the next phase, leading to more accurate mean estimates of the rewards in the subsequent phase. This positive reinforcement cycle is critical to reducing the number of uniform exploration rounds and closely matching our lower bound.
♻ ☆ Variational formulations of ODE-Net as a mean-field optimal control problem and existence results
This paper presents a mathematical analysis of ODE-Net, a continuum model of deep neural networks (DNNs). In recent years, Machine Learning researchers have introduced ideas of replacing the deep structure of DNNs with ODEs as a continuum limit. These studies regard the "learning" of ODE-Net as the minimization of a "loss" constrained by a parametric ODE. Although the existence of a minimizer for this minimization problem needs to be assumed, only a few studies have investigated its existence analytically in detail. In the present paper, the existence of a minimizer is discussed based on a formulation of ODE-Net as a measure-theoretic mean-field optimal control problem. The existence result is proved when a neural network, which describes a vector field of ODE-Net, is linear with respect to learnable parameters. The proof employs the measure-theoretic formulation combined with the direct method of Calculus of Variations. Secondly, an idealized minimization problem is proposed to remove the above linearity assumption. Such a problem is inspired by a kinetic regularization associated with the Benamou--Brenier formula and universal approximation theorems for neural networks. The proofs of these existence results use variational methods, differential equations, and mean-field optimal control theory. They will stand for a new analytic way to investigate the learning process of deep neural networks.
comment: 33 pages
♻ ☆ PlaNeRF: SVD Unsupervised 3D Plane Regularization for NeRF Large-Scale Scene Reconstruction
Neural Radiance Fields (NeRF) enable 3D scene reconstruction from 2D images and camera poses for Novel View Synthesis (NVS). Although NeRF can produce photorealistic results, it often suffers from overfitting to training views, leading to poor geometry reconstruction, especially in low-texture areas. This limitation restricts many important applications which require accurate geometry, such as extrapolated NVS, HD mapping and scene editing. To address this limitation, we propose a new method to improve NeRF's 3D structure using only RGB images and semantic maps. Our approach introduces a novel plane regularization based on Singular Value Decomposition (SVD), that does not rely on any geometric prior. In addition, we leverage the Structural Similarity Index Measure (SSIM) in our loss design to properly initialize the volumetric representation of NeRF. Quantitative and qualitative results show that our method outperforms popular regularization approaches in accurate geometry reconstruction for large-scale outdoor scenes and achieves SoTA rendering quality on the KITTI-360 NVS benchmark.
comment: 14 pages, 7 figures
♻ ☆ I Know What You Trained Last Summer: A Survey on Stealing Machine Learning Models and Defences
Machine Learning-as-a-Service (MLaaS) has become a widespread paradigm, making even the most complex machine learning models available for clients via e.g. a pay-per-query principle. This allows users to avoid time-consuming processes of data collection, hyperparameter tuning, and model training. However, by giving their customers access to the (predictions of their) models, MLaaS providers endanger their intellectual property, such as sensitive training data, optimised hyperparameters, or learned model parameters. Adversaries can create a copy of the model with (almost) identical behavior using the the prediction labels only. While many variants of this attack have been described, only scattered defence strategies have been proposed, addressing isolated threats. This raises the necessity for a thorough systematisation of the field of model stealing, to arrive at a comprehensive understanding why these attacks are successful, and how they could be holistically defended against. We address this by categorising and comparing model stealing attacks, assessing their performance, and exploring corresponding defence techniques in different settings. We propose a taxonomy for attack and defence approaches, and provide guidelines on how to select the right attack or defence strategy based on the goal and available resources. Finally, we analyse which defences are rendered less effective by current attack strategies.
comment: Accepted at ACM Computing Surveys, 2023: https://doi.org/10.1145/3595292
♻ ☆ Seizing Serendipity: Exploiting the Value of Past Success in Off-Policy Actor-Critic
Learning high-quality Q-value functions plays a key role in the success of many modern off-policy deep reinforcement learning (RL) algorithms. Previous works focus on addressing the value overestimation issue, an outcome of adopting function approximators and off-policy learning. Deviating from the common viewpoint, we observe that Q-values are indeed underestimated in the latter stage of the RL training process, primarily related to the use of inferior actions from the current policy in Bellman updates as compared to the more optimal action samples in the replay buffer. We hypothesize that this long-neglected phenomenon potentially hinders policy learning and reduces sample efficiency. Our insight to address this issue is to incorporate sufficient exploitation of past successes while maintaining exploration optimism. We propose the Blended Exploitation and Exploration (BEE) operator, a simple yet effective approach that updates Q-value using both historical best-performing actions and the current policy. The instantiations of our method in both model-free and model-based settings outperform state-of-the-art methods in various continuous control tasks and achieve strong performance in failure-prone scenarios and real-world robot tasks.
♻ ☆ Denise: Deep Robust Principal Component Analysis for Positive Semidefinite Matrices
The robust PCA of covariance matrices plays an essential role when isolating key explanatory features. The currently available methods for performing such a low-rank plus sparse decomposition are matrix specific, meaning, those algorithms must re-run for every new matrix. Since these algorithms are computationally expensive, it is preferable to learn and store a function that nearly instantaneously performs this decomposition when evaluated. Therefore, we introduce Denise, a deep learning-based algorithm for robust PCA of covariance matrices, or more generally, of symmetric positive semidefinite matrices, which learns precisely such a function. Theoretical guarantees for Denise are provided. These include a novel universal approximation theorem adapted to our geometric deep learning problem and convergence to an optimal solution to the learning problem. Our experiments show that Denise matches state-of-the-art performance in terms of decomposition quality, while being approximately $2000\times$ faster than the state-of-the-art, principal component pursuit (PCP), and $200 \times$ faster than the current speed-optimized method, fast PCP.
♻ ☆ ArrayFlex: A Systolic Array Architecture with Configurable Transparent Pipelining DATE 2023
Convolutional Neural Networks (CNNs) are the state-of-the-art solution for many deep learning applications. For maximum scalability, their computation should combine high performance and energy efficiency. In practice, the convolutions of each CNN layer are mapped to a matrix multiplication that includes all input features and kernels of each layer and is computed using a systolic array. In this work, we focus on the design of a systolic array with configurable pipeline with the goal to select an optimal pipeline configuration for each CNN layer. The proposed systolic array, called ArrayFlex, can operate in normal, or in shallow pipeline mode, thus balancing the execution time in cycles and the operating clock frequency. By selecting the appropriate pipeline configuration per CNN layer, ArrayFlex reduces the inference latency of state-of-the-art CNNs by 11%, on average, as compared to a traditional fixed-pipeline systolic array. Most importantly, this result is achieved while using 13%-23% less power, for the same applications, thus offering a combined energy-delay-product efficiency between 1.4x and 1.8x.
comment: DATE 2023
♻ ☆ In Search of Insights, Not Magic Bullets: Towards Demystification of the Model Selection Dilemma in Heterogeneous Treatment Effect Estimation
Personalized treatment effect estimates are often of interest in high-stakes applications -- thus, before deploying a model estimating such effects in practice, one needs to be sure that the best candidate from the ever-growing machine learning toolbox for this task was chosen. Unfortunately, due to the absence of counterfactual information in practice, it is usually not possible to rely on standard validation metrics for doing so, leading to a well-known model selection dilemma in the treatment effect estimation literature. While some solutions have recently been investigated, systematic understanding of the strengths and weaknesses of different model selection criteria is still lacking. In this paper, instead of attempting to declare a global `winner', we therefore empirically investigate success- and failure modes of different selection criteria. We highlight that there is a complex interplay between selection strategies, candidate estimators and the data used for comparing them, and provide interesting insights into the relative (dis)advantages of different criteria alongside desiderata for the design of further illuminating empirical studies in this context.
comment: To appear in the Proceedings of the 40th International Conference on Machine Learning, Honolulu, Hawaii, USA. PMLR 202, 2023
♻ ☆ A Kernel-Based View of Language Model Fine-Tuning ICML 2023
It has become standard to solve NLP tasks by fine-tuning pre-trained language models (LMs), especially in low-data settings. There is minimal theoretical understanding of empirical success, e.g., why fine-tuning a model with $10^8$ or more parameters on a couple dozen training points does not result in overfitting. We investigate whether the Neural Tangent Kernel (NTK) - which originated as a model to study the gradient descent dynamics of infinitely wide networks with suitable random initialization - describes fine-tuning of pre-trained LMs. This study was inspired by the decent performance of NTK for computer vision tasks (Wei et al., 2022). We extend the NTK formalism to Adam and use Tensor Programs (Yang, 2020) to characterize conditions under which the NTK lens may describe fine-tuning updates to pre-trained language models. Extensive experiments on 14 NLP tasks validate our theory and show that formulating the downstream task as a masked word prediction problem through prompting often induces kernel-based dynamics during fine-tuning. Finally, we use this kernel view to propose an explanation for the success of parameter-efficient subspace-based fine-tuning methods.
comment: Accepted at ICML 2023. Code and pre-computed kernels are publicly available at https://github.com/princeton-nlp/LM-Kernel-FT
♻ ☆ RHINO: Rotated DETR with Dynamic Denoising via Hungarian Matching for Oriented Object Detection
With the publication of DINO, a variant of the Detection Transformer (DETR), Detection Transformers are breaking the record in the object detection benchmark with the merits of their end-to-end design and scalability. However, the extension of DETR to oriented object detection has not been thoroughly studied although more benefits from its end-to-end architecture are expected such as removing NMS and anchor-related costs. In this paper, we propose a first strong DINO-based baseline for oriented object detection. We found that straightforward employment of DETRs for oriented object detection does not guarantee non-duplicate prediction, and propose a simple cost to mitigate this. Furthermore, we introduce a $\textit{dynamic denoising}$ strategy that uses Hungarian matching to filter redundant noised queries and $\textit{query alignment}$ to preserve matching consistency between Transformer decoder layers. Our proposed model outperforms previous rotated DETRs and other counterparts, achieving state-of-the-art performance in DOTA-v1.0/v1.5/v2.0, and DIOR-R benchmarks.
comment: State-of-the-art rotated object detector in DOTA v1.0/v1.5/v2.0 and DIOR-R at the time of publication
♻ ☆ Safe Peeling for L0-Regularized Least-Squares with supplementary material
We introduce a new methodology dubbed ``safe peeling'' to accelerate the resolution of L0-regularized least-squares problems via a Branch-and-Bound (BnB) algorithm. Our procedure enables to tighten the convex relaxation considered at each node of the BnB decision tree and therefore potentially allows for more aggressive pruning. Numerical simulations show that our proposed methodology leads to significant gains in terms of number of nodes explored and overall solving time.s show that our proposed methodology leads to significant gains in terms of number of nodes explored and overall solving time.
♻ ☆ I Prefer not to Say: Protecting User Consent in Models with Optional Personal Data NeurIPS 2022
We examine machine learning models in a setup where individuals have the choice to share optional personal information with a decision-making system, as seen in modern insurance pricing models. Some users consent to their data being used whereas others object and keep their data undisclosed. In this work, we show that the decision not to share data can be considered as information in itself that should be protected to respect users' privacy. This observation raises the overlooked problem of how to ensure that users who protect their personal data do not suffer any disadvantages as a result. To address this problem, we formalize protection requirements for models which only use the information for which active user consent was obtained. This excludes implicit information contained in the decision to share data or not. We offer the first solution to this problem by proposing the notion of Protected User Consent (PUC), which we prove to be loss-optimal under our protection requirement. To learn PUC-compliant models, we devise a model-agnostic data augmentation strategy with finite sample convergence guarantees. Finally, we analyze the implications of PUC on a variety of challenging real-world datasets, tasks, and models.
comment: Updated Version. v1 accepted at NeurIPS 2022 Workshop on Algorithmic Fairness through the Lens of Causality and Privacy (AFCP)
♻ ☆ Context-aware multi-head self-attentional neural network model for next location prediction
Accurate activity location prediction is a crucial component of many mobility applications and is particularly required to develop personalized, sustainable transportation systems. Despite the widespread adoption of deep learning models, next location prediction models lack a comprehensive discussion and integration of mobility-related spatio-temporal contexts. Here, we utilize a multi-head self-attentional (MHSA) neural network that learns location transition patterns from historical location visits, their visit time and activity duration, as well as their surrounding land use functions, to infer an individual's next location. Specifically, we adopt point-of-interest data and latent Dirichlet allocation for representing locations' land use contexts at multiple spatial scales, generate embedding vectors of the spatio-temporal features, and learn to predict the next location with an MHSA network. Through experiments on two large-scale GNSS tracking datasets, we demonstrate that the proposed model outperforms other state-of-the-art prediction models, and reveal the contribution of various spatio-temporal contexts to the model's performance. Moreover, we find that the model trained on population data achieves higher prediction performance with fewer parameters than individual-level models due to learning from collective movement patterns. We also reveal mobility conducted in the recent past and one week before has the largest influence on the current prediction, showing that learning from a subset of the historical mobility is sufficient to obtain an accurate location prediction result. We believe that the proposed model is vital for context-aware mobility prediction. The gained insights will help to understand location prediction models and promote their implementation for mobility applications.
comment: Updated figures and added more descriptions in Appendix
♻ ☆ Global Context Vision Transformers ICML 2023
We propose global context vision transformer (GC ViT), a novel architecture that enhances parameter and compute utilization for computer vision. Our method leverages global context self-attention modules, joint with standard local self-attention, to effectively and efficiently model both long and short-range spatial interactions, without the need for expensive operations such as computing attention masks or shifting local windows. In addition, we address the lack of the inductive bias in ViTs, and propose to leverage a modified fused inverted residual blocks in our architecture. Our proposed GC ViT achieves state-of-the-art results across image classification, object detection and semantic segmentation tasks. On ImageNet-1K dataset for classification, the variants of GC ViT with 51M, 90M and 201M parameters achieve 84.3%, 85.0% and 85.7% Top-1 accuracy, respectively, at 224 image resolution and without any pre-training, hence surpassing comparably-sized prior art such as CNN-based ConvNeXt and ViT-based MaxViT and Swin Transformer by a large margin. Pre-trained GC ViT backbones in downstream tasks of object detection, instance segmentation, and semantic segmentation using MS COCO and ADE20K datasets outperform prior work consistently. Specifically, GC ViT with a 4-scale DINO detection head achieves a box AP of 58.3 on MS COCO dataset.
comment: Accepted to ICML 2023
♻ ☆ Graph Neural Rough Differential Equations for Traffic Forecasting
Traffic forecasting is one of the most popular spatio-temporal tasks in the field of machine learning. A prevalent approach in the field is to combine graph convolutional networks and recurrent neural networks for the spatio-temporal processing. There has been fierce competition and many novel methods have been proposed. In this paper, we present the method of spatio-temporal graph neural rough differential equation (STG-NRDE). Neural rough differential equations (NRDEs) are a breakthrough concept for processing time-series data. Their main concept is to use the log-signature transform to convert a time-series sample into a relatively shorter series of feature vectors. We extend the concept and design two NRDEs: one for the temporal processing and the other for the spatial processing. After that, we combine them into a single framework. We conduct experiments with 6 benchmark datasets and 27 baselines. STG-NRDE shows the best accuracy in all cases, outperforming all those 27 baselines by non-trivial margins.
comment: Accepted to ACM Transactions on Intelligent Systems and Technology (ACM TIST). arXiv admin note: substantial text overlap with arXiv:2112.03558
♻ ☆ DL-DRL: A double-level deep reinforcement learning approach for large-scale task scheduling of multi-UAV
Exploiting unmanned aerial vehicles (UAVs) to execute tasks is gaining growing popularity recently. To solve the underlying task scheduling problem, the deep reinforcement learning (DRL) based methods demonstrate notable advantage over the conventional heuristics as they rely less on hand-engineered rules. However, their decision space will become prohibitively huge as the problem scales up, thus deteriorating the computation efficiency. To alleviate this issue, we propose a double-level deep reinforcement learning (DL-DRL) approach based on a divide and conquer framework (DCF), where we decompose the task scheduling of multi-UAV into task allocation and route planning. Particularly, we design an encoder-decoder structured policy network in our upper-level DRL model to allocate the tasks to different UAVs, and we exploit another attention based policy network in our lower-level DRL model to construct the route for each UAV, with the objective to maximize the number of executed tasks given the maximum flight distance of the UAV. To effectively train the two models, we design an interactive training strategy (ITS), which includes pre-training, intensive training and alternate training. Experimental results show that our DL-DRL performs favorably against the learning-based and conventional baselines including the OR-Tools, in terms of solution quality and computation efficiency. We also verify the generalization performance of our approach by applying it to larger sizes of up to 1000 tasks. Moreover, we also show via an ablation study that our ITS can help achieve a balance between the performance and training efficiency.
comment: 13 pages, 7 figures
♻ ☆ Cold PAWS: Unsupervised class discovery and addressing the cold-start problem for semi-supervised learning
In many machine learning applications, labeling datasets can be an arduous and time-consuming task. Although research has shown that semi-supervised learning techniques can achieve high accuracy with very few labels within the field of computer vision, little attention has been given to how images within a dataset should be selected for labeling. In this paper, we propose a novel approach based on well-established self-supervised learning, clustering, and manifold learning techniques that address this challenge of selecting an informative image subset to label in the first instance, which is known as the cold-start or unsupervised selective labelling problem. We test our approach using several publicly available datasets, namely CIFAR10, Imagenette, DeepWeeds, and EuroSAT, and observe improved performance with both supervised and semi-supervised learning strategies when our label selection strategy is used, in comparison to random sampling. We also obtain superior performance for the datasets considered with a much simpler approach compared to other methods in the literature.
♻ ☆ Deep learning for diffusion in porous media
We adopt convolutional neural networks (CNN) to predict the basic properties of the porous media. Two different media types are considered: one mimics the sand packings, and the other mimics the systems derived from the extracellular space of biological tissues. The Lattice Boltzmann Method is used to obtain the labeled data necessary for performing supervised learning. We distinguish two tasks. In the first, networks based on the analysis of the system's geometry predict porosity and effective diffusion coefficient. In the second, networks reconstruct the concentration map. In the first task, we propose two types of CNN models: the C-Net and the encoder part of the U-Net. Both networks are modified by adding a self-normalization module [Graczyk \textit{et al.}, Sci Rep 12, 10583 (2022)]. The models predict with reasonable accuracy but only within the data type, they are trained on. For instance, the model trained on sand packings-like samples overshoots or undershoots for biological-like samples. In the second task, we propose the usage of the U-Net architecture. It accurately reconstructs the concentration fields. In contrast to the first task, the network trained on one data type works well for the other. For instance, the model trained on sand packings-like samples works perfectly on biological-like samples. Eventually, for both types of the data, we fit exponents in the Archie's law to find tortuosity that is used to describe the dependence of the effective diffusion on porosity.
comment: 15 pages, 17 figures, to appear in Sci. Rep
♻ ☆ Entropy-driven Unsupervised Keypoint Representation Learning in Videos ICML 2023
Extracting informative representations from videos is fundamental for effectively learning various downstream tasks. We present a novel approach for unsupervised learning of meaningful representations from videos, leveraging the concept of image spatial entropy (ISE) that quantifies the per-pixel information in an image. We argue that \textit{local entropy} of pixel neighborhoods and their temporal evolution create valuable intrinsic supervisory signals for learning prominent features. Building on this idea, we abstract visual features into a concise representation of keypoints that act as dynamic information transmitters, and design a deep learning model that learns, purely unsupervised, spatially and temporally consistent representations \textit{directly} from video frames. Two original information-theoretic losses, computed from local entropy, guide our model to discover consistent keypoint representations; a loss that maximizes the spatial information covered by the keypoints and a loss that optimizes the keypoints' information transportation over time. We compare our keypoint representation to strong baselines for various downstream tasks, \eg, learning object dynamics. Our empirical results show superior performance for our information-driven keypoints that resolve challenges like attendance to static and dynamic objects or objects abruptly entering and leaving the scene.
comment: 29 pages, 14 figures, Accepted at ICML 2023
Multimedia 6
☆ Pivotuner: automatic real-time pure intonation and microtonal modulation
Pivotuner is a VST3/AU MIDI effect plugin that automatically tunes note data in an adaptive pure intonation, in real time. Where previously pure intonation was out of reach for most musicians due to difficulty and impracticality, Pivotuner enables it to be achieved easily and straightforwardly by using novel yet simple algorithms. This may lead to more widespread exploration of pure intonation for a larger and more diverse crowd of musicians! This paper includes a review of prior systems for adaptive pure intonation systems, including Hermode Tuning/Kontakt Dynamic Pure Tuning and Just Intonation. The paper introduces the notion of an adaptive tuning center and how it serves as a flexible underlying concept for multiple tuning algorithms, as well as extensions to offer greater control for performers, including pitch and tuning center locking and resetting, and gradual interpolation between equal temperament and pure intonation. The paper then showcases some pieces which use Pivotuner effectively, then discusses areas for future exploration within Pivotuner's feature set, and plans for future development.
comment: 5 pages, associated files and additional information available at https://www.dmitrivolkov.com/projects/pivotuner/
☆ Emotion-Conditioned Melody Harmonization with Hierarchical Variational Autoencoder
Existing melody harmonization models have made great progress in improving the quality of generated harmonies, but most of them ignored the emotions beneath the music. Meanwhile, the variability of harmonies generated by previous methods is insufficient. To solve these problems, we propose a novel LSTM-based Hierarchical Variational Auto-Encoder (LHVAE) to investigate the influence of emotional conditions on melody harmonization, while improving the quality of generated harmonies and capturing the abundant variability of chord progressions. Specifically, LHVAE incorporates latent variables and emotional conditions at different levels (piece- and bar-level) to model the global and local music properties. Additionally, we introduce an attention-based melody context vector at each step to better learn the correspondence between melodies and harmonies. Experimental results of the objective evaluation show that our proposed model outperforms other LSTM-based models. Through subjective evaluation, we conclude that only altering the chords hardly changes the overall emotion of the music. The qualitative analysis demonstrates the ability of our model to generate variable harmonies.
comment: Accepted by IEEE SMC 2023
☆ SGAT4PASS: Spherical Geometry-Aware Transformer for PAnoramic Semantic Segmentation IJCAI 2023
As an important and challenging problem in computer vision, PAnoramic Semantic Segmentation (PASS) gives complete scene perception based on an ultra-wide angle of view. Usually, prevalent PASS methods with 2D panoramic image input focus on solving image distortions but lack consideration of the 3D properties of original $360^{\circ}$ data. Therefore, their performance will drop a lot when inputting panoramic images with the 3D disturbance. To be more robust to 3D disturbance, we propose our Spherical Geometry-Aware Transformer for PAnoramic Semantic Segmentation (SGAT4PASS), considering 3D spherical geometry knowledge. Specifically, a spherical geometry-aware framework is proposed for PASS. It includes three modules, i.e., spherical geometry-aware image projection, spherical deformable patch embedding, and a panorama-aware loss, which takes input images with 3D disturbance into account, adds a spherical geometry-aware constraint on the existing deformable patch embedding, and indicates the pixel density of original $360^{\circ}$ data, respectively. Experimental results on Stanford2D3D Panoramic datasets show that SGAT4PASS significantly improves performance and robustness, with approximately a 2% increase in mIoU, and when small 3D disturbances occur in the data, the stability of our performance is improved by an order of magnitude. Our code and supplementary material are available at https://github.com/TencentARC/SGAT4PASS.
comment: Accepted by IJCAI 2023
☆ Computational Technologies for Fashion Recommendation: A Survey
Fashion recommendation is a key research field in computational fashion research and has attracted considerable interest in the computer vision, multimedia, and information retrieval communities in recent years. Due to the great demand for applications, various fashion recommendation tasks, such as personalized fashion product recommendation, complementary (mix-and-match) recommendation, and outfit recommendation, have been posed and explored in the literature. The continuing research attention and advances impel us to look back and in-depth into the field for a better understanding. In this paper, we comprehensively review recent research efforts on fashion recommendation from a technological perspective. We first introduce fashion recommendation at a macro level and analyse its characteristics and differences with general recommendation tasks. We then clearly categorize different fashion recommendation efforts into several sub-tasks and focus on each sub-task in terms of its problem formulation, research focus, state-of-the-art methods, and limitations. We also summarize the datasets proposed in the literature for use in fashion recommendation studies to give readers a brief illustration. Finally, we discuss several promising directions for future research in this field. Overall, this survey systematically reviews the development of fashion recommendation research. It also discusses the current limitations and gaps between academic research and the real needs of the fashion industry. In the process, we offer a deep insight into how the fashion industry could benefit from fashion recommendation technologies. the computational technologies of fashion recommendation.
☆ Systematic Analysis of Music Representations from BERT
There have been numerous attempts to represent raw data as numerical vectors that effectively capture semantic and contextual information. However, in the field of symbolic music, previous works have attempted to validate their music embeddings by observing the performance improvement of various fine-tuning tasks. In this work, we directly analyze embeddings from BERT and BERT with contrastive learning trained on bar-level MIDI, inspecting their musical information that can be obtained from MIDI events. We observe that the embeddings exhibit distinct characteristics of information depending on the contrastive objectives and the choice of layers. Our code is available at https://github.com/sjhan91/MusicBERT.
♻ ☆ Towards Unified Text-based Person Retrieval: A Large-scale Multi-Attribute and Language Search Benchmark
In this paper, we introduce a large Multi-Attribute and Language Search dataset for text-based person retrieval, called MALS, and explore the feasibility of performing pre-training on both attribute recognition and image-text matching tasks in one stone. In particular, MALS contains 1,510,330 image-text pairs, which is about 37.5 times larger than prevailing CUHK-PEDES, and all images are annotated with 27 attributes. Considering the privacy concerns and annotation costs, we leverage the off-the-shelf diffusion models to generate the dataset. To verify the feasibility of learning from the generated data, we develop a new joint Attribute Prompt Learning and Text Matching Learning (APTM) framework, considering the shared knowledge between attribute and text. As the name implies, APTM contains an attribute prompt learning stream and a text matching learning stream. (1) The attribute prompt learning leverages the attribute prompts for image-attribute alignment, which enhances the text matching learning. (2) The text matching learning facilitates the representation learning on fine-grained details, and in turn, boosts the attribute prompt learning. Extensive experiments validate the effectiveness of the pre-training on MALS, achieving state-of-the-art retrieval performance via APTM on three challenging real-world benchmarks. In particular, APTM achieves a consistent improvement of +6.60%, +7.39%, and +15.90% Recall@1 accuracy on CUHK-PEDES, ICFG-PEDES, and RSTPReid datasets by a clear margin, respectively.
Computation and Language 100
☆ RepoBench: Benchmarking Repository-Level Code Auto-Completion Systems
Large Language Models (LLMs) have greatly advanced code auto-completion systems, with a potential for substantial productivity enhancements for developers. However, current benchmarks mainly focus on single-file tasks, leaving an assessment gap for more complex, real-world, multi-file programming scenarios. To fill this gap, we introduce RepoBench, a new benchmark specifically designed for evaluating repository-level code auto-completion systems. RepoBench consists of three interconnected evaluation tasks: RepoBench-R (Retrieval), RepoBench-C (Code Completion), and RepoBench-P (Pipeline). Each task respectively measures the system's ability to retrieve the most relevant code snippets from other files as cross-file context, predict the next line of code with cross-file and in-file context, and handle complex tasks that require a combination of both retrieval and next-line prediction. RepoBench aims to facilitate a more complete comparison of performance and encouraging continuous improvement in auto-completion systems. RepoBench is publicly available at https://github.com/Leolty/repobench.
☆ Is ChatGPT a Good Teacher Coach? Measuring Zero-Shot Performance For Scoring and Providing Actionable Insights on Classroom Instruction
Coaching, which involves classroom observation and expert feedback, is a widespread and fundamental part of teacher training. However, the majority of teachers do not have access to consistent, high quality coaching due to limited resources and access to expertise. We explore whether generative AI could become a cost-effective complement to expert feedback by serving as an automated teacher coach. In doing so, we propose three teacher coaching tasks for generative AI: (A) scoring transcript segments based on classroom observation instruments, (B) identifying highlights and missed opportunities for good instructional strategies, and (C) providing actionable suggestions for eliciting more student reasoning. We recruit expert math teachers to evaluate the zero-shot performance of ChatGPT on each of these tasks for elementary math classroom transcripts. Our results reveal that ChatGPT generates responses that are relevant to improving instruction, but they are often not novel or insightful. For example, 82% of the model's suggestions point to places in the transcript where the teacher is already implementing that suggestion. Our work highlights the challenges of producing insightful, novel and truthful feedback for teachers while paving the way for future research to address these obstacles and improve the capacity of generative AI to coach teachers.
comment: In the Proceedings of Innovative Use of NLP for Building Educational Applications 2023; The code and model outputs are open-sourced here: https://github.com/rosewang2008/zero-shot-teacher-feedback
☆ Sequential Monte Carlo Steering of Large Language Models using Probabilistic Programs
Even after fine-tuning and reinforcement learning, large language models (LLMs) can be difficult, if not impossible, to control reliably with prompts alone. We propose a new inference-time approach to enforcing syntactic and semantic constraints on the outputs of LLMs, called sequential Monte Carlo (SMC) steering. The key idea is to specify language generation tasks as posterior inference problems in a class of discrete probabilistic sequence models, and replace standard decoding with sequential Monte Carlo inference. For a computational cost similar to that of beam search, SMC can steer LLMs to solve diverse tasks, including infilling, generation under syntactic constraints, and prompt intersection. To facilitate experimentation with SMC steering, we present a probabilistic programming library, LLaMPPL (https://github.com/probcomp/LLaMPPL), for concisely specifying new generation tasks as language model probabilistic programs, and automating steering of LLaMA-family Transformers.
☆ Machine Learning and Statistical Approaches to Measuring Similarity of Political Parties
Mapping political party systems to metric policy spaces is one of the major methodological problems in political science. At present, in most political science project this task is performed by domain experts relying on purely qualitative assessments, with all the attendant problems of subjectivity and labor intensiveness. We consider how advances in natural language processing, including large transformer-based language models, can be applied to solve that issue. We apply a number of texts similarity measures to party political programs, analyze how they correlate with each other, and -- in the absence of a satisfactory benchmark -- evaluate them against other measures, including those based on expert surveys, voting records, electoral patterns, and candidate networks. Finally, we consider the prospects of relying on those methods to correct, supplement, and eventually replace expert judgments.
☆ SpQR: A Sparse-Quantized Representation for Near-Lossless LLM Weight Compression
Recent advances in large language model (LLM) pretraining have led to high-quality LLMs with impressive abilities. By compressing such LLMs via quantization to 3-4 bits per parameter, they can fit into memory-limited devices such as laptops and mobile phones, enabling personalized use. However, quantization down to 3-4 bits per parameter usually leads to moderate-to-high accuracy losses, especially for smaller models in the 1-10B parameter range, which are well-suited for edge deployments. To address this accuracy issue, we introduce the Sparse-Quantized Representation (SpQR), a new compressed format and quantization technique which enables for the first time near-lossless compression of LLMs across model scales, while reaching similar compression levels to previous methods. SpQR works by identifying and isolating outlier weights, which cause particularly-large quantization errors, and storing them in higher precision, while compressing all other weights to 3-4 bits, and achieves relative accuracy losses of less than 1% in perplexity for highly-accurate LLaMA and Falcon LLMs. This makes it possible to run 33B parameter LLM on a single 24 GB consumer GPU without any performance degradation at 15% speedup thus making powerful LLMs available to consumer without any downsides. SpQR comes with efficient algorithms for both encoding weights into its format, as well as decoding them efficiently at runtime. Specifically, we provide an efficient GPU inference algorithm for SpQR which yields faster inference than 16-bit baselines at similar accuracy, while enabling memory compression gains of more than 4x.
comment: Extended preprint
☆ Interactive Editing for Text Summarization
Summarizing lengthy documents is a common and essential task in our daily lives. Although recent advancements in neural summarization models can assist in crafting general-purpose summaries, human writers often have specific requirements that call for a more customized approach. To address this need, we introduce REVISE (Refinement and Editing via Iterative Summarization Enhancement), an innovative framework designed to facilitate iterative editing and refinement of draft summaries by human writers. Within our framework, writers can effortlessly modify unsatisfactory segments at any location or length and provide optional starting phrases -- our system will generate coherent alternatives that seamlessly integrate with the existing summary. At its core, REVISE incorporates a modified fill-in-the-middle model with the encoder-decoder architecture while developing novel evaluation metrics tailored for the summarization task. In essence, our framework empowers users to create high-quality, personalized summaries by effectively harnessing both human expertise and AI capabilities, ultimately transforming the summarization process into a truly collaborative and adaptive experience.
☆ Structured Voronoi Sampling
Recently, there has been a growing interest in the development of gradient-based sampling algorithms for text generation, especially in the context of controlled generation. However, there exists a lack of theoretically grounded and principled approaches for this task. In this paper, we take an important step toward building a principled approach for sampling from language models with gradient-based methods. We use discrete distributions given by language models to define densities and develop an algorithm based on Hamiltonian Monte Carlo to sample from them. We name our gradient-based technique Structured Voronoi Sampling (SVS). In an experimental setup where the reference distribution is known, we show that the empirical distribution of SVS samples is closer to the reference distribution compared to alternative sampling schemes. Furthermore, in a controlled generation task, SVS is able to generate fluent and diverse samples while following the control targets significantly better than other methods.
☆ Analyzing Syntactic Generalization Capacity of Pre-trained Language Models on Japanese Honorific Conversion ACL2023
Using Japanese honorifics is challenging because it requires not only knowledge of the grammatical rules but also contextual information, such as social relationships. It remains unclear whether pre-trained large language models (LLMs) can flexibly handle Japanese honorifics like humans. To analyze this, we introduce an honorific conversion task that considers social relationships among people mentioned in a conversation. We construct a Japanese honorifics dataset from problem templates of various sentence structures to investigate the syntactic generalization capacity of GPT-3, one of the leading LLMs, on this task under two settings: fine-tuning and prompt learning. Our results showed that the fine-tuned GPT-3 performed better in a context-aware honorific conversion task than the prompt-based one. The fine-tuned model demonstrated overall syntactic generalizability towards compound honorific sentences, except when tested with the data involving direct speech.
comment: To appear in the Proceedings of the 12th Joint Conference on Lexical and Computational Semantics (*SEM2023) with ACL2023
☆ Benchmarking Large Language Models on CMExam -- A Comprehensive Chinese Medical Exam Dataset
Recent advancements in large language models (LLMs) have transformed the field of question answering (QA). However, evaluating LLMs in the medical field is challenging due to the lack of standardized and comprehensive datasets. To address this gap, we introduce CMExam, sourced from the Chinese National Medical Licensing Examination. CMExam consists of 60K+ multiple-choice questions for standardized and objective evaluations, as well as solution explanations for model reasoning evaluation in an open-ended manner. For in-depth analyses of LLMs, we invited medical professionals to label five additional question-wise annotations, including disease groups, clinical departments, medical disciplines, areas of competency, and question difficulty levels. Alongside the dataset, we further conducted thorough experiments with representative LLMs and QA algorithms on CMExam. The results show that GPT-4 had the best accuracy of 61.5% and a weighted F1 score of 0.616. These results highlight a great disparity when compared to human accuracy, which stood at 71.6%. For explanation tasks, while LLMs could generate relevant reasoning and demonstrate improved performance after finetuning, they fall short of a desired standard, indicating ample room for improvement. To the best of our knowledge, CMExam is the first Chinese medical exam dataset to provide comprehensive medical annotations. The experiments and findings of LLM evaluation also provide valuable insights into the challenges and potential solutions in developing Chinese medical QA systems and LLM evaluation pipelines. The dataset and relevant code are available at https://github.com/williamliujl/CMExam.
☆ PokemonChat: Auditing ChatGPT for Pokémon Universe Knowledge
The recently released ChatGPT model demonstrates unprecedented capabilities in zero-shot question-answering. In this work, we probe ChatGPT for its conversational understanding and introduce a conversational framework (protocol) that can be adopted in future studies. The Pok\'emon universe serves as an ideal testing ground for auditing ChatGPT's reasoning capabilities due to its closed world assumption. After bringing ChatGPT's background knowledge (on the Pok\'emon universe) to light, we test its reasoning process when using these concepts in battle scenarios. We then evaluate its ability to acquire new knowledge and include it in its reasoning process. Our ultimate goal is to assess ChatGPT's ability to generalize, combine features, and to acquire and reason over newly introduced knowledge from human feedback. We find that ChatGPT has prior knowledge of the Pokemon universe, which can reason upon in battle scenarios to a great extent, even when new information is introduced. The model performs better with collaborative feedback and if there is an initial phase of information retrieval, but also hallucinates occasionally and is susceptible to adversarial attacks.
☆ PolyVoice: Language Models for Speech to Speech Translation
We propose PolyVoice, a language model-based framework for speech-to-speech translation (S2ST) system. Our framework consists of two language models: a translation language model and a speech synthesis language model. We use discretized speech units, which are generated in a fully unsupervised way, and thus our framework can be used for unwritten languages. For the speech synthesis part, we adopt the existing VALL-E X approach and build a unit-based audio language model. This grants our framework the ability to preserve the voice characteristics and the speaking style of the original speech. We examine our system on Chinese $\rightarrow$ English and English $\rightarrow$ Spanish pairs. Experimental results show that our system can generate speech with high translation quality and audio quality. Speech samples are available at https://speechtranslation.github.io/polyvoice.
☆ KNOW How to Make Up Your Mind! Adversarially Detecting and Alleviating Inconsistencies in Natural Language Explanations ACL 2023
While recent works have been considerably improving the quality of the natural language explanations (NLEs) generated by a model to justify its predictions, there is very limited research in detecting and alleviating inconsistencies among generated NLEs. In this work, we leverage external knowledge bases to significantly improve on an existing adversarial attack for detecting inconsistent NLEs. We apply our attack to high-performing NLE models and show that models with higher NLE quality do not necessarily generate fewer inconsistencies. Moreover, we propose an off-the-shelf mitigation method to alleviate inconsistencies by grounding the model into external background knowledge. Our method decreases the inconsistencies of previous high-performing NLE models as detected by our attack.
comment: Short paper, ACL 2023
☆ Which Argumentative Aspects of Hate Speech in Social Media can be reliably identified?
With the increasing diversity of use cases of large language models, a more informative treatment of texts seems necessary. An argumentative analysis could foster a more reasoned usage of chatbots, text completion mechanisms or other applications. However, it is unclear which aspects of argumentation can be reliably identified and integrated in language models. In this paper, we present an empirical assessment of the reliability with which different argumentative aspects can be automatically identified in hate speech in social media. We have enriched the Hateval corpus (Basile et al. 2019) with a manual annotation of some argumentative components, adapted from Wagemans (2016)'s Periodic Table of Arguments. We show that some components can be identified with reasonable reliability. For those that present a high error ratio, we analyze the patterns of disagreement between expert annotators and errors in automatic procedures, and we propose adaptations of those categories that can be more reliably reproduced.
comment: 9 Pages plus reference and appendix
☆ A Simple and Flexible Modeling for Mental Disorder Detection by Learning from Clinical Questionnaires ACL 2023
Social media is one of the most highly sought resources for analyzing characteristics of the language by its users. In particular, many researchers utilized various linguistic features of mental health problems from social media. However, existing approaches to detecting mental disorders face critical challenges, such as the scarcity of high-quality data or the trade-off between addressing the complexity of models and presenting interpretable results grounded in expert domain knowledge. To address these challenges, we design a simple but flexible model that preserves domain-based interpretability. We propose a novel approach that captures the semantic meanings directly from the text and compares them to symptom-related descriptions. Experimental results demonstrate that our model outperforms relevant baselines on various mental disorder detection tasks. Our detailed analysis shows that the proposed model is effective at leveraging domain knowledge, transferable to other mental disorders, and providing interpretable detection results.
comment: ACL 2023, 15 pages, 11 tables, 4 figures
☆ MidMed: Towards Mixed-Type Dialogues for Medical Consultation ACL 2023
Most medical dialogue systems assume that patients have clear goals (medicine querying, surgical operation querying, etc.) before medical consultation. However, in many real scenarios, due to the lack of medical knowledge, it is usually difficult for patients to determine clear goals with all necessary slots. In this paper, we identify this challenge as how to construct medical consultation dialogue systems to help patients clarify their goals. To mitigate this challenge, we propose a novel task and create a human-to-human mixed-type medical consultation dialogue corpus, termed MidMed, covering five dialogue types: task-oriented dialogue for diagnosis, recommendation, knowledge-grounded dialogue, QA, and chitchat. MidMed covers four departments (otorhinolaryngology, ophthalmology, skin, and digestive system), with 8,175 dialogues. Furthermore, we build baselines on MidMed and propose an instruction-guiding medical dialogue generation framework, termed InsMed, to address this task. Experimental results show the effectiveness of InsMed.
comment: Accepted by ACL 2023 Main conference. First two authors contributed equally to this work
☆ Second Language Acquisition of Neural Language Models
With the success of neural language models (LMs), their language acquisition has gained much attention. This work sheds light on the second language (L2) acquisition of LMs, while previous work has typically explored their first language (L1) acquisition. Specifically, we trained bilingual LMs with a scenario similar to human L2 acquisition and analyzed their cross-lingual transfer from linguistic perspectives. Our exploratory experiments demonstrated that the L1 pretraining accelerated their linguistic generalization in L2, and language transfer configurations (e.g., the L1 choice, and presence of parallel texts) substantially affected their generalizations. These clarify their (non-)human-like L2 acquisition in particular aspects.
☆ SelfEvolve: A Code Evolution Framework via Large Language Models
Large language models (LLMs) have already revolutionized code generation, after being pretrained on publicly available code data. However, while various methods have been proposed to augment LLMs with retrieved knowledge and enhance the quality of code generation, the performance of these retrieval-based methods is limited by the strength of the retrievers used. In addition, while LLMs show great emergent ability, they still struggle to produce the correct code in one turn. To address these challenges, we propose a novel two-step pipeline, called \autoknow, that leverages LLMs as both knowledge providers and self-reflective programmers. Unlike retrieval-based methods, \autoknow~obtains the knowledge from input prompts and generates intermediate code based on the generated knowledge. After that, \autoknow~asks LLM to act as an expert programmer to perform debugging for the generated code. This is achieved by receiving the error message from the interpreter, without requiring special test cases for correctness verification. We evaluate \autoknow~on three code generation datasets, including DS-1000 for data science code, HumanEval for software engineering code, and TransCoder for C++-to-Python translation. Our empirical experiments show that \autoknow~outperforms strong baselines by a significant margin on all datasets. We also conduct exhaustive analytical experiments to validate the effectiveness of the two stages of \autoknow, and find that both are superior to other prompting-based methods. Further scalability analysis demonstrates that \autoknow~can be adapted to other more advanced models, such as GPT-4, and bring consistent efficacy improvement.
☆ N-Shot Benchmarking of Whisper on Diverse Arabic Speech Recognition INTERSPEECH 2023
Whisper, the recently developed multilingual weakly supervised model, is reported to perform well on multiple speech recognition benchmarks in both monolingual and multilingual settings. However, it is not clear how Whisper would fare under diverse conditions even on languages it was evaluated on such as Arabic. In this work, we address this gap by comprehensively evaluating Whisper on several varieties of Arabic speech for the ASR task. Our evaluation covers most publicly available Arabic speech data and is performed under n-shot (zero-, few-, and full) finetuning. We also investigate the robustness of Whisper under completely novel conditions, such as in dialect-accented standard Arabic and in unseen dialects for which we develop evaluation data. Our experiments show that although Whisper zero-shot outperforms fully finetuned XLS-R models on all datasets, its performance deteriorates significantly in the zero-shot setting for five unseen dialects (i.e., Algeria, Jordan, Palestine, UAE, and Yemen).
comment: 4 pages, INTERSPEECH 2023
☆ Gen-IR @ SIGIR 2023: The First Workshop on Generative Information Retrieval SIGIR 23
Generative information retrieval (IR) has experienced substantial growth across multiple research communities (e.g., information retrieval, computer vision, natural language processing, and machine learning), and has been highly visible in the popular press. Theoretical, empirical, and actual user-facing products have been released that retrieve documents (via generation) or directly generate answers given an input request. We would like to investigate whether end-to-end generative models are just another trend or, as some claim, a paradigm change for IR. This necessitates new metrics, theoretical grounding, evaluation methods, task definitions, models, user interfaces, etc. The goal of this workshop (https://coda.io/@sigir/gen-ir) is to focus on previously explored Generative IR techniques like document retrieval and direct Grounded Answer Generation, while also offering a venue for the discussion and exploration of how Generative IR can be applied to new domains like recommendation systems, summarization, etc. The format of the workshop is interactive, including roundtable and keynote sessions and tends to avoid the one-sided dialogue of a mini-conference.
comment: Accepted SIGIR 23 workshop
☆ DecompX: Explaining Transformers Decisions by Propagating Token Decomposition ACL 2023
An emerging solution for explaining Transformer-based models is to use vector-based analysis on how the representations are formed. However, providing a faithful vector-based explanation for a multi-layer model could be challenging in three aspects: (1) Incorporating all components into the analysis, (2) Aggregating the layer dynamics to determine the information flow and mixture throughout the entire model, and (3) Identifying the connection between the vector-based analysis and the model's predictions. In this paper, we present DecompX to tackle these challenges. DecompX is based on the construction of decomposed token representations and their successive propagation throughout the model without mixing them in between layers. Additionally, our proposal provides multiple advantages over existing solutions for its inclusion of all encoder components (especially nonlinear feed-forward networks) and the classification head. The former allows acquiring precise vectors while the latter transforms the decomposition into meaningful prediction-based values, eliminating the need for norm- or summation-based vector aggregation. According to the standard faithfulness evaluations, DecompX consistently outperforms existing gradient-based and vector-based approaches on various datasets. Our code is available at https://github.com/mohsenfayyaz/DecompX.
comment: Accepted to ACL 2023 (main conference)
☆ Text-To-KG Alignment: Comparing Current Methods on Classification Tasks ATC
In contrast to large text corpora, knowledge graphs (KG) provide dense and structured representations of factual information. This makes them attractive for systems that supplement or ground the knowledge found in pre-trained language models with an external knowledge source. This has especially been the case for classification tasks, where recent work has focused on creating pipeline models that retrieve information from KGs like ConceptNet as additional context. Many of these models consist of multiple components, and although they differ in the number and nature of these parts, they all have in common that for some given text query, they attempt to identify and retrieve a relevant subgraph from the KG. Due to the noise and idiosyncrasies often found in KGs, it is not known how current methods compare to a scenario where the aligned subgraph is completely relevant to the query. In this work, we try to bridge this knowledge gap by reviewing current approaches to text-to-KG alignment and evaluating them on two datasets where manually created graphs are available, providing insights into the effectiveness of current methods.
comment: Camera ready version for MATCHING workshop at ACL 2023
☆ On "Scientific Debt" in NLP: A Case for More Rigour in Language Model Pre-Training Research ACL 2023
This evidence-based position paper critiques current research practices within the language model pre-training literature. Despite rapid recent progress afforded by increasingly better pre-trained language models (PLMs), current PLM research practices often conflate different possible sources of model improvement, without conducting proper ablation studies and principled comparisons between different models under comparable conditions. These practices (i) leave us ill-equipped to understand which pre-training approaches should be used under what circumstances; (ii) impede reproducibility and credit assignment; and (iii) render it difficult to understand: "How exactly does each factor contribute to the progress that we have today?" We provide a case in point by revisiting the success of BERT over its baselines, ELMo and GPT-1, and demonstrate how -- under comparable conditions where the baselines are tuned to a similar extent -- these baselines (and even-simpler variants thereof) can, in fact, achieve competitive or better performance than BERT. These findings demonstrate how disentangling different factors of model improvements can lead to valuable new insights. We conclude with recommendations for how to encourage and incentivize this line of work, and accelerate progress towards a better and more systematic understanding of what factors drive the progress of our foundation models today.
comment: Accepted at ACL 2023
☆ Leveraging Large Language Models for Topic Classification in the Domain of Public Affairs ICDAR 2023
The analysis of public affairs documents is crucial for citizens as it promotes transparency, accountability, and informed decision-making. It allows citizens to understand government policies, participate in public discourse, and hold representatives accountable. This is crucial, and sometimes a matter of life or death, for companies whose operation depend on certain regulations. Large Language Models (LLMs) have the potential to greatly enhance the analysis of public affairs documents by effectively processing and understanding the complex language used in such documents. In this work, we analyze the performance of LLMs in classifying public affairs documents. As a natural multi-label task, the classification of these documents presents important challenges. In this work, we use a regex-powered tool to collect a database of public affairs documents with more than 33K samples and 22.5M tokens. Our experiments assess the performance of 4 different Spanish LLMs to classify up to 30 different topics in the data in different configurations. The results shows that LLMs can be of great use to process domain-specific documents, such as those in the domain of public affairs.
comment: Accepted in ICDAR 2023 Workshop on Automatic Domain-Adapted and Personalized Document Analysis
☆ Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding
We present Video-LLaMA, a multi-modal framework that empowers Large Language Models (LLMs) with the capability of understanding both visual and auditory content in the video. Video-LLaMA bootstraps cross-modal training from the frozen pre-trained visual \& audio encoders and the frozen LLMs. Unlike previous vision- LLMs that focus on static image comprehensions such as MiniGPT-4~\citep{zhu2023minigpt} and LLaVA~\citep{liu2023visualit}, Video-LLaMA tackles two challenges in video understanding: (1) capturing the temporal changes in visual scenes, (2) integrating audio-visual signals. For the first challenge, we propose Video Q-former to extend the pre-trained image encoder to a video encoder and introduce a video-to-text generation task to learn video-language correspondence. For the second challenge, we leverage ImageBind~\citep{girdhar2023imagebind} as the pre-trained audio encoder which performs exceptionally well in aligning different modalities to a common embedding space. And then introduce an Audio Q-former to learn auditory query tokens. To align the output of both visual \& audio encoder with LLM's embedding space, we train Video-LLaMA on a large-scale vision caption dataset and a hign-quantity vision-instruction-tuning dataset. We found Video-LLaMA showcases the ability to perceive and comprehend video content, generating meaningful responses that are grounded in the visual and auditory information present in the videos. This highlights the potential of Video-LLaMA as a promising prototype for audio-visual AI assistants. Our code, pre-trained model, and demo are available at \url{https://github.com/DAMO-NLP-SG/Video-LLaMA}.
comment: Technical Report
☆ Improving Conversational Recommendation Systems via Counterfactual Data Simulation KDD 2023
Conversational recommender systems (CRSs) aim to provide recommendation services via natural language conversations. Although a number of approaches have been proposed for developing capable CRSs, they typically rely on sufficient training data for training. Since it is difficult to annotate recommendation-oriented dialogue datasets, existing CRS approaches often suffer from the issue of insufficient training due to the scarcity of training data. To address this issue, in this paper, we propose a CounterFactual data simulation approach for CRS, named CFCRS, to alleviate the issue of data scarcity in CRSs. Our approach is developed based on the framework of counterfactual data augmentation, which gradually incorporates the rewriting to the user preference from a real dialogue without interfering with the entire conversation flow. To develop our approach, we characterize user preference and organize the conversation flow by the entities involved in the dialogue, and design a multi-stage recommendation dialogue simulator based on a conversation flow language model. Under the guidance of the learned user preference and dialogue schema, the flow language model can produce reasonable, coherent conversation flows, which can be further realized into complete dialogues. Based on the simulator, we perform the intervention at the representations of the interacted entities of target users, and design an adversarial training method with a curriculum schedule that can gradually optimize the data augmentation strategy. Extensive experiments show that our approach can consistently boost the performance of several competitive CRSs, and outperform other data augmentation methods, especially when the training data is limited. Our code is publicly available at https://github.com/RUCAIBox/CFCRS.
comment: Accepted by KDD 2023. Code: https://github.com/RUCAIBox/CFCRS
☆ Learning to Substitute Spans towards Improving Compositional Generalization ACL 2023
Despite the rising prevalence of neural sequence models, recent empirical evidences suggest their deficiency in compositional generalization. One of the current de-facto solutions to this problem is compositional data augmentation, aiming to incur additional compositional inductive bias. Nonetheless, the improvement offered by existing handcrafted augmentation strategies is limited when successful systematic generalization of neural sequence models requires multi-grained compositional bias (i.e., not limited to either lexical or structural biases only) or differentiation of training sequences in an imbalanced difficulty distribution. To address the two challenges, we first propose a novel compositional augmentation strategy dubbed \textbf{Span} \textbf{Sub}stitution (SpanSub) that enables multi-grained composition of substantial substructures in the whole training set. Over and above that, we introduce the \textbf{L}earning \textbf{to} \textbf{S}ubstitute \textbf{S}pan (L2S2) framework which empowers the learning of span substitution probabilities in SpanSub in an end-to-end manner by maximizing the loss of neural sequence models, so as to outweigh those challenging compositions with elusive concepts and novel surroundings. Our empirical results on three standard compositional generalization benchmarks, including SCAN, COGS and GeoQuery (with an improvement of at most 66.5\%, 10.3\%, 1.2\%, respectively), demonstrate the superiority of SpanSub, %the learning framework L2S2 and their combination.
comment: accepted by ACL 2023
☆ UNIDECOR: A Unified Deception Corpus for Cross-Corpus Deception Detection
Verbal deception has been studied in psychology, forensics, and computational linguistics for a variety of reasons, like understanding behaviour patterns, identifying false testimonies, and detecting deception in online communication. Varying motivations across research fields lead to differences in the domain choices to study and in the conceptualization of deception, making it hard to compare models and build robust deception detection systems for a given language. With this paper, we improve this situation by surveying available English deception datasets which include domains like social media reviews, court testimonials, opinion statements on specific topics, and deceptive dialogues from online strategy games. We consolidate these datasets into a single unified corpus. Based on this resource, we conduct a correlation analysis of linguistic cues of deception across datasets to understand the differences and perform cross-corpus modeling experiments which show that a cross-domain generalization is challenging to achieve. The unified deception corpus (UNIDECOR) can be obtained from https://www.ims.uni-stuttgart.de/data/unidecor.
☆ Enhancing Language Representation with Constructional Information for Natural Language Understanding ACL 2023
Natural language understanding (NLU) is an essential branch of natural language processing, which relies on representations generated by pre-trained language models (PLMs). However, PLMs primarily focus on acquiring lexico-semantic information, while they may be unable to adequately handle the meaning of constructions. To address this issue, we introduce construction grammar (CxG), which highlights the pairings of form and meaning, to enrich language representation. We adopt usage-based construction grammar as the basis of our work, which is highly compatible with statistical models such as PLMs. Then a HyCxG framework is proposed to enhance language representation through a three-stage solution. First, all constructions are extracted from sentences via a slot-constraints approach. As constructions can overlap with each other, bringing redundancy and imbalance, we formulate the conditional max coverage problem for selecting the discriminative constructions. Finally, we propose a relational hypergraph attention network to acquire representation from constructional information by capturing high-order word interactions among constructions. Extensive experiments demonstrate the superiority of the proposed model on a variety of NLU tasks.
comment: Long paper, accepted at the ACL 2023
☆ Modeling Human-like Concept Learning with Bayesian Inference over Natural Language
We model learning of abstract symbolic concepts by performing Bayesian inference over utterances in natural language. For efficient inference, we use a large language model as a proposal distribution. We fit a prior to human data to better model human learners, and evaluate on both generative and logical concepts.
☆ MCTS: A Multi-Reference Chinese Text Simplification Dataset
Text simplification aims to make the text easier to understand by applying rewriting transformations. There has been very little research on Chinese text simplification for a long time. The lack of generic evaluation data is an essential reason for this phenomenon. In this paper, we introduce MCTS, a multi-reference Chinese text simplification dataset. We describe the annotation process of the dataset and provide a detailed analysis of it. Furthermore, we evaluate the performance of some unsupervised methods and advanced large language models. We hope to build a basic understanding of Chinese text simplification through the foundational work and provide references for future research. We release our data at https://github.com/blcuicall/mcts.
☆ Exploring the Relationship between Alignment and Cross-lingual Transfer in Multilingual Transformers
Without any explicit cross-lingual training data, multilingual language models can achieve cross-lingual transfer. One common way to improve this transfer is to perform realignment steps before fine-tuning, i.e., to train the model to build similar representations for pairs of words from translated sentences. But such realignment methods were found to not always improve results across languages and tasks, which raises the question of whether aligned representations are truly beneficial for cross-lingual transfer. We provide evidence that alignment is actually significantly correlated with cross-lingual transfer across languages, models and random seeds. We show that fine-tuning can have a significant impact on alignment, depending mainly on the downstream task and the model. Finally, we show that realignment can, in some instances, improve cross-lingual transfer, and we identify conditions in which realignment methods provide significant improvements. Namely, we find that realignment works better on tasks for which alignment is correlated with cross-lingual transfer when generalizing to a distant language and with smaller models, as well as when using a bilingual dictionary rather than FastAlign to extract realignment pairs. For example, for POS-tagging, between English and Arabic, realignment can bring a +15.8 accuracy improvement on distilmBERT, even outperforming XLM-R Large by 1.7. We thus advocate for further research on realignment methods for smaller multilingual models as an alternative to scaling.
☆ German CheXpert Chest X-ray Radiology Report Labeler
This study aimed to develop an algorithm to automatically extract annotations for chest X-ray classification models from German thoracic radiology reports. An automatic label extraction model was designed based on the CheXpert architecture, and a web-based annotation interface was created for iterative improvements. Results showed that automated label extraction can reduce time spent on manual labeling and improve overall modeling performance. The model trained on automatically extracted labels performed competitively to manually labeled data and strongly outperformed the model trained on publicly available data.
☆ Identifying the style by a qualified reader on a short fragment of generated poetry
Style is an important concept in today's challenges in natural language generating. After the success in the field of image style transfer, the task of text style transfer became actual and attractive. Researchers are also interested in the tasks of style reproducing in generation of the poetic text. Evaluation of style reproducing in natural poetry generation remains a problem. I used 3 character-based LSTM-models to work with style reproducing assessment. All three models were trained on the corpus of texts by famous Russian-speaking poets. Samples were shown to the assessors and 4 answer options were offered, the style of which poet this sample reproduces. In addition, the assessors were asked how well they were familiar with the work of the poet they had named. Students studying history of literature were the assessors, 94 answers were received. It has appeared that accuracy of definition of style increases if the assessor can quote the poet by heart. Each model showed at least 0.7 macro-average accuracy. The experiment showed that it is better to involve a professional rather than a naive reader in the evaluation of style in the tasks of poetry generation, while lstm models are good at reproducing the style of Russian poets even on a limited training corpus.
comment: 6 pages, 2 tables
☆ Cross-Lingual Transfer with Target Language-Ready Task Adapters ACL 2023
Adapters have emerged as a modular and parameter-efficient approach to (zero-shot) cross-lingual transfer. The established MAD-X framework employs separate language and task adapters which can be arbitrarily combined to perform the transfer of any task to any target language. Subsequently, BAD-X, an extension of the MAD-X framework, achieves improved transfer at the cost of MAD-X's modularity by creating "bilingual" adapters specific to the source-target language pair. In this work, we aim to take the best of both worlds by (i) fine-tuning task adapters adapted to the target language(s) (so-called "target language-ready" (TLR) adapters) to maintain high transfer performance, but (ii) without sacrificing the highly modular design of MAD-X. The main idea of "target language-ready" adapters is to resolve the training-vs-inference discrepancy of MAD-X: the task adapter "sees" the target language adapter for the very first time during inference, and thus might not be fully compatible with it. We address this mismatch by exposing the task adapter to the target language adapter during training, and empirically validate several variants of the idea: in the simplest form, we alternate between using the source and target language adapters during task adapter training, which can be generalized to cycling over any set of language adapters. We evaluate different TLR-based transfer configurations with varying degrees of generality across a suite of standard cross-lingual benchmarks, and find that the most general (and thus most modular) configuration consistently outperforms MAD-X and BAD-X on most tasks and languages.
comment: Accepted to Findings of ACL 2023
☆ PULSAR: Pre-training with Extracted Healthcare Terms for Summarising Patients' Problems and Data Augmentation with Black-box Large Language Models ACL 2023
Medical progress notes play a crucial role in documenting a patient's hospital journey, including his or her condition, treatment plan, and any updates for healthcare providers. Automatic summarisation of a patient's problems in the form of a problem list can aid stakeholders in understanding a patient's condition, reducing workload and cognitive bias. BioNLP 2023 Shared Task 1A focuses on generating a list of diagnoses and problems from the provider's progress notes during hospitalisation. In this paper, we introduce our proposed approach to this task, which integrates two complementary components. One component employs large language models (LLMs) for data augmentation; the other is an abstractive summarisation LLM with a novel pre-training objective for generating the patients' problems summarised as a list. Our approach was ranked second among all submissions to the shared task. The performance of our model on the development and test datasets shows that our approach is more robust on unknown data, with an improvement of up to 3.1 points over the same size of the larger model.
comment: Accepted by ACL 2023's workshop BioNLP 2023
☆ Multiple output samples for each input in a single-output Gaussian process
The standard Gaussian Process (GP) only considers a single output sample per input in the training set. Datasets for subjective tasks, such as spoken language assessment, may be annotated with output labels from multiple human raters per input. This paper proposes to generalise the GP to allow for these multiple output samples in the training set, and thus make use of available output uncertainty information. This differs from a multi-output GP, as all output samples are from the same task here. The output density function is formulated to be the joint likelihood of observing all output samples, and latent variables are not repeated to reduce computation cost. The test set predictions are inferred similarly to a standard GP, with a difference being in the optimised hyper-parameters. This is evaluated on speechocean762, showing that it allows the GP to compute a test set output distribution that is more similar to the collection of reference outputs from the multiple human raters.
☆ Orca: Progressive Learning from Complex Explanation Traces of GPT-4
Recent research has focused on enhancing the capability of smaller models through imitation learning, drawing on the outputs generated by large foundation models (LFMs). A number of issues impact the quality of these models, ranging from limited imitation signals from shallow LFM outputs; small scale homogeneous training data; and most notably a lack of rigorous evaluation resulting in overestimating the small model's capability as they tend to learn to imitate the style, but not the reasoning process of LFMs. To address these challenges, we develop Orca (We are working with our legal team to publicly release a diff of the model weights in accordance with LLaMA's release policy to be published at https://aka.ms/orca-lm), a 13-billion parameter model that learns to imitate the reasoning process of LFMs. Orca learns from rich signals from GPT-4 including explanation traces; step-by-step thought processes; and other complex instructions, guided by teacher assistance from ChatGPT. To promote this progressive learning, we tap into large-scale and diverse imitation data with judicious sampling and selection. Orca surpasses conventional state-of-the-art instruction-tuned models such as Vicuna-13B by more than 100% in complex zero-shot reasoning benchmarks like Big-Bench Hard (BBH) and 42% on AGIEval. Moreover, Orca reaches parity with ChatGPT on the BBH benchmark and shows competitive performance (4 pts gap with optimized system message) in professional and academic examinations like the SAT, LSAT, GRE, and GMAT, both in zero-shot settings without CoT; while trailing behind GPT-4. Our research indicates that learning from step-by-step explanations, whether these are generated by humans or more advanced AI models, is a promising direction to improve model capabilities and skills.
☆ CELDA: Leveraging Black-box Language Model as Enhanced Classifier without Labels ACL 2023
Utilizing language models (LMs) without internal access is becoming an attractive paradigm in the field of NLP as many cutting-edge LMs are released through APIs and boast a massive scale. The de-facto method in this type of black-box scenario is known as prompting, which has shown progressive performance enhancements in situations where data labels are scarce or unavailable. Despite their efficacy, they still fall short in comparison to fully supervised counterparts and are generally brittle to slight modifications. In this paper, we propose Clustering-enhanced Linear Discriminative Analysis, a novel approach that improves the text classification accuracy with a very weak-supervision signal (i.e., name of the labels). Our framework draws a precise decision boundary without accessing weights or gradients of the LM model or data labels. The core ideas of CELDA are twofold: (1) extracting a refined pseudo-labeled dataset from an unlabeled dataset, and (2) training a lightweight and robust model on the top of LM, which learns an accurate decision boundary from an extracted noisy dataset. Throughout in-depth investigations on various datasets, we demonstrated that CELDA reaches new state-of-the-art in weakly-supervised text classification and narrows the gap with a fully-supervised model. Additionally, our proposed methodology can be applied universally to any LM and has the potential to scale to larger models, making it a more viable option for utilizing large LMs.
comment: ACL 2023
☆ End-to-End Word-Level Pronunciation Assessment with MASK Pre-training
Pronunciation assessment is a major challenge in the computer-aided pronunciation training system, especially at the word (phoneme)-level. To obtain word (phoneme)-level scores, current methods usually rely on aligning components to obtain acoustic features of each word (phoneme), which limits the performance of assessment to the accuracy of alignments. Therefore, to address this problem, we propose a simple yet effective method, namely \underline{M}asked pre-training for \underline{P}ronunciation \underline{A}ssessment (MPA). Specifically, by incorporating a mask-predict strategy, our MPA supports end-to-end training without leveraging any aligning components and can solve misalignment issues to a large extent during prediction. Furthermore, we design two evaluation strategies to enable our model to conduct assessments in both unsupervised and supervised settings. Experimental results on SpeechOcean762 dataset demonstrate that MPA could achieve better performance than previous methods, without any explicit alignment. In spite of this, MPA still has some limitations, such as requiring more inference time and reference text. They expect to be addressed in future work.
comment: Accepted by InterSpeech 2023
☆ BeAts: Bengali Speech Acts Recognition using Multimodal Attention Fusion INTERSPEECH 2023
Spoken languages often utilise intonation, rhythm, intensity, and structure, to communicate intention, which can be interpreted differently depending on the rhythm of speech of their utterance. These speech acts provide the foundation of communication and are unique in expression to the language. Recent advancements in attention-based models, demonstrating their ability to learn powerful representations from multilingual datasets, have performed well in speech tasks and are ideal to model specific tasks in low resource languages. Here, we develop a novel multimodal approach combining two models, wav2vec2.0 for audio and MarianMT for text translation, by using multimodal attention fusion to predict speech acts in our prepared Bengali speech corpus. We also show that our model BeAts ($\underline{\textbf{Be}}$ngali speech acts recognition using Multimodal $\underline{\textbf{At}}$tention Fu$\underline{\textbf{s}}$ion) significantly outperforms both the unimodal baseline using only speech data and a simpler bimodal fusion using both speech and text data. Project page: https://soumitri2001.github.io/BeAts
comment: Accepted at INTERSPEECH 2023
☆ Joint Pre-training and Local Re-training: Transferable Representation Learning on Multi-source Knowledge Graphs KDD
In this paper, we present the ``joint pre-training and local re-training'' framework for learning and applying multi-source knowledge graph (KG) embeddings. We are motivated by the fact that different KGs contain complementary information to improve KG embeddings and downstream tasks. We pre-train a large teacher KG embedding model over linked multi-source KGs and distill knowledge to train a student model for a task-specific KG. To enable knowledge transfer across different KGs, we use entity alignment to build a linked subgraph for connecting the pre-trained KGs and the target KG. The linked subgraph is re-trained for three-level knowledge distillation from the teacher to the student, i.e., feature knowledge distillation, network knowledge distillation, and prediction knowledge distillation, to generate more expressive embeddings. The teacher model can be reused for different target KGs and tasks without having to train from scratch. We conduct extensive experiments to demonstrate the effectiveness and efficiency of our framework.
comment: Accepted in the 29th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD 2023)
☆ Improving Grammar-based Sequence-to-Sequence Modeling with Decomposition and Constraints ACL 2023
Neural QCFG is a grammar-based sequence-tosequence (seq2seq) model with strong inductive biases on hierarchical structures. It excels in interpretability and generalization but suffers from expensive inference. In this paper, we study two low-rank variants of Neural QCFG for faster inference with different trade-offs between efficiency and expressiveness. Furthermore, utilizing the symbolic interface provided by the grammar, we introduce two soft constraints over tree hierarchy and source coverage. We experiment with various datasets and find that our models outperform vanilla Neural QCFG in most settings.
comment: ACL 2023
☆ Colexifications for Bootstrapping Cross-lingual Datasets: The Case of Phonology, Concreteness, and Affectiveness
Colexification refers to the linguistic phenomenon where a single lexical form is used to convey multiple meanings. By studying cross-lingual colexifications, researchers have gained valuable insights into fields such as psycholinguistics and cognitive sciences [Jackson et al.,2019]. While several multilingual colexification datasets exist, there is untapped potential in using this information to bootstrap datasets across such semantic features. In this paper, we aim to demonstrate how colexifications can be leveraged to create such cross-lingual datasets. We showcase curation procedures which result in a dataset covering 142 languages across 21 language families across the world. The dataset includes ratings of concreteness and affectiveness, mapped with phonemes and phonological features. We further analyze the dataset along different dimensions to demonstrate potential of the proposed procedures in facilitating further interdisciplinary research in psychology, cognitive science, and multilingual natural language processing (NLP). Based on initial investigations, we observe that i) colexifications that are closer in concreteness/affectiveness are more likely to colexify; ii) certain initial/last phonemes are significantly correlated with concreteness/affectiveness intra language families, such as /k/ as the initial phoneme in both Turkic and Tai-Kadai correlated with concreteness, and /p/ in Dravidian and Sino-Tibetan correlated with Valence; iii) the type-to-token ratio (TTR) of phonemes are positively correlated with concreteness across several language families, while the length of phoneme segments are negatively correlated with concreteness; iv) certain phonological features are negatively correlated with concreteness across languages. The dataset is made public online for further research.
comment: 13 pages, 4 figures, accepted to SIGMORPHON 2023
☆ Do-GOOD: Towards Distribution Shift Evaluation for Pre-Trained Visual Document Understanding Models SIGIR 2023
Numerous pre-training techniques for visual document understanding (VDU) have recently shown substantial improvements in performance across a wide range of document tasks. However, these pre-trained VDU models cannot guarantee continued success when the distribution of test data differs from the distribution of training data. In this paper, to investigate how robust existing pre-trained VDU models are to various distribution shifts, we first develop an out-of-distribution (OOD) benchmark termed Do-GOOD for the fine-Grained analysis on Document image-related tasks specifically. The Do-GOOD benchmark defines the underlying mechanisms that result in different distribution shifts and contains 9 OOD datasets covering 3 VDU related tasks, e.g., document information extraction, classification and question answering. We then evaluate the robustness and perform a fine-grained analysis of 5 latest VDU pre-trained models and 2 typical OOD generalization algorithms on these OOD datasets. Results from the experiments demonstrate that there is a significant performance gap between the in-distribution (ID) and OOD settings for document images, and that fine-grained analysis of distribution shifts can reveal the brittle nature of existing pre-trained VDU models and OOD generalization algorithms. The code and datasets for our Do-GOOD benchmark can be found at https://github.com/MAEHCM/Do-GOOD.
comment: SIGIR 2023. The code and datasets for our Do-GOOD benchmark can be found at https://github.com/MAEHCM/Do-GOOD
☆ What Makes Entities Similar? A Similarity Flooding Perspective for Multi-sourced Knowledge Graph Embeddings ICML 2023
Joint representation learning over multi-sourced knowledge graphs (KGs) yields transferable and expressive embeddings that improve downstream tasks. Entity alignment (EA) is a critical step in this process. Despite recent considerable research progress in embedding-based EA, how it works remains to be explored. In this paper, we provide a similarity flooding perspective to explain existing translation-based and aggregation-based EA models. We prove that the embedding learning process of these models actually seeks a fixpoint of pairwise similarities between entities. We also provide experimental evidence to support our theoretical analysis. We propose two simple but effective methods inspired by the fixpoint computation in similarity flooding, and demonstrate their effectiveness on benchmark datasets. Our work bridges the gap between recent embedding-based models and the conventional similarity flooding algorithm. It would improve our understanding of and increase our faith in embedding-based EA.
comment: Accepted in the 40th International Conference on Machine Learning (ICML 2023)
☆ Building Resilient SMEs: Harnessing Large Language Models for Cyber Security in Australia
The escalating digitalisation of our lives and enterprises has led to a parallel growth in the complexity and frequency of cyber-attacks. Small and medium-sized enterprises (SMEs), particularly in Australia, are experiencing increased vulnerability to cyber threats, posing a significant challenge to the nation's cyber security landscape. Embracing transformative technologies such as Artificial Intelligence (AI), Machine Learning (ML) and Large Language Models (LLMs) can potentially strengthen cyber security policies for Australian SMEs. However, their practical application, advantages, and limitations remain underexplored, with prior research mainly focusing on large corporations. This study aims to address this gap by providing a comprehensive understanding of the potential role of LLMs in enhancing cyber security policies for Australian SMEs. Employing a mixed-methods study design, this research includes a literature review, qualitative analysis of SME case studies, and a quantitative assessment of LLM performance metrics in cyber security applications. The findings highlight the promising potential of LLMs across various performance criteria, including relevance, accuracy, and applicability, though gaps remain in areas such as completeness and clarity. The study underlines the importance of integrating human expertise with LLM technology and refining model development to address these limitations. By proposing a robust conceptual framework guiding the effective adoption of LLMs, this research aims to contribute to a safer and more resilient cyber environment for Australian SMEs, enabling sustainable growth and competitiveness in the digital era.
comment: 8 pages, 1 figure
☆ Early Rumor Detection Using Neural Hawkes Process with a New Benchmark Dataset NAACL 2022
Little attention has been paid on \underline{EA}rly \underline{R}umor \underline{D}etection (EARD), and EARD performance was evaluated inappropriately on a few datasets where the actual early-stage information is largely missing. To reverse such situation, we construct BEARD, a new \underline{B}enchmark dataset for \underline{EARD}, based on claims from fact-checking websites by trying to gather as many early relevant posts as possible. We also propose HEARD, a novel model based on neural \underline{H}awkes process for \underline{EARD}, which can guide a generic rumor detection model to make timely, accurate and stable predictions. Experiments show that HEARD achieves effective EARD performance on two commonly used general rumor detection datasets and our BEARD dataset.
comment: Accepted at NAACL 2022
☆ A Novel Interpretable and Generalizable Re-synchronization Model for Cued Speech based on a Multi-Cuer Corpus INTERSPEECH2023
Cued Speech (CS) is a multi-modal visual coding system combining lip reading with several hand cues at the phonetic level to make the spoken language visible to the hearing impaired. Previous studies solved asynchronous problems between lip and hand movements by a cuer\footnote{The people who perform Cued Speech are called the cuer.}-dependent piecewise linear model for English and French CS. In this work, we innovatively propose three statistical measure on the lip stream to build an interpretable and generalizable model for predicting hand preceding time (HPT), which achieves cuer-independent by a proper normalization. Particularly, we build the first Mandarin CS corpus comprising annotated videos from five speakers including three normal and two hearing impaired individuals. Consequently, we show that the hand preceding phenomenon exists in Mandarin CS production with significant differences between normal and hearing impaired people. Extensive experiments demonstrate that our model outperforms the baseline and the previous state-of-the-art methods.
comment: 5 pages, 4 figures, Accepted to INTERSPEECH2023
☆ Graph-Aware Language Model Pre-Training on a Large Graph Corpus Can Help Multiple Graph Applications KDD 2023
Model pre-training on large text corpora has been demonstrated effective for various downstream applications in the NLP domain. In the graph mining domain, a similar analogy can be drawn for pre-training graph models on large graphs in the hope of benefiting downstream graph applications, which has also been explored by several recent studies. However, no existing study has ever investigated the pre-training of text plus graph models on large heterogeneous graphs with abundant textual information (a.k.a. large graph corpora) and then fine-tuning the model on different related downstream applications with different graph schemas. To address this problem, we propose a framework of graph-aware language model pre-training (GALM) on a large graph corpus, which incorporates large language models and graph neural networks, and a variety of fine-tuning methods on downstream applications. We conduct extensive experiments on Amazon's real internal datasets and large public datasets. Comprehensive empirical results and in-depth analysis demonstrate the effectiveness of our proposed methods along with lessons learned.
comment: To be published in the KDD 2023 proceedings as a full paper
☆ Cross-Lingual Transfer Learning for Phrase Break Prediction with Multilingual Language Model INTERSPEECH 2023
Phrase break prediction is a crucial task for improving the prosody naturalness of a text-to-speech (TTS) system. However, most proposed phrase break prediction models are monolingual, trained exclusively on a large amount of labeled data. In this paper, we address this issue for low-resource languages with limited labeled data using cross-lingual transfer. We investigate the effectiveness of zero-shot and few-shot cross-lingual transfer for phrase break prediction using a pre-trained multilingual language model. We use manually collected datasets in four Indo-European languages: one high-resource language and three with limited resources. Our findings demonstrate that cross-lingual transfer learning can be a particularly effective approach, especially in the few-shot setting, for improving performance in low-resource languages. This suggests that cross-lingual transfer can be inexpensive and effective for developing TTS front-end in resource-poor languages.
comment: Accepted by INTERSPEECH 2023
Prompt to be Consistent is Better than Self-Consistent? Few-Shot and Zero-Shot Fact Verification with Pre-trained Language Models ACL 2023
Few-shot or zero-shot fact verification only relies on a few or no labeled training examples. In this paper, we propose a novel method called ProToCo, to \underline{Pro}mpt pre-trained language models (PLMs) \underline{To} be \underline{Co}nsistent, for improving the factuality assessment capability of PLMs in the few-shot and zero-shot settings. Given a claim-evidence pair, ProToCo generates multiple variants of the claim with different relations and frames a simple consistency mechanism as constraints for making compatible predictions across these variants. We update PLMs by using parameter-efficient fine-tuning (PEFT), leading to more accurate predictions in few-shot and zero-shot fact verification tasks. Our experiments on three public verification datasets show that ProToCo significantly outperforms state-of-the-art few-shot fact verification baselines. With a small number of unlabeled instances, ProToCo also outperforms the strong zero-shot learner T0 on zero-shot verification. Compared to large PLMs using in-context learning (ICL) method, ProToCo outperforms OPT-30B and the Self-Consistency-enabled OPT-6.7B model in both few- and zero-shot settings.
comment: Accepted as ACL 2023 Findings
☆ LLM-Blender: Ensembling Large Language Models with Pairwise Ranking and Generative Fusion ACL 2023
We present LLM-Blender, an ensembling framework designed to attain consistently superior performance by leveraging the diverse strengths of multiple open-source large language models (LLMs). Our framework consists of two modules: PairRanker and GenFuser, addressing the observation that optimal LLMs for different examples can significantly vary. PairRanker employs a specialized pairwise comparison method to distinguish subtle differences between candidate outputs. It jointly encodes the input text and a pair of candidates, using cross-attention encoders to determine the superior one. Our results demonstrate that PairRanker exhibits the highest correlation with ChatGPT-based ranking. Then, GenFuser aims to merge the top-ranked candidates, generating an improved output by capitalizing on their strengths and mitigating their weaknesses. To facilitate large-scale evaluation, we introduce a benchmark dataset, MixInstruct, which is a mixture of multiple instruction datasets featuring oracle pairwise comparisons. Our LLM-Blender significantly outperform individual LLMs and baseline methods across various metrics, establishing a substantial performance gap.
comment: ACL 2023 (Main conference). Project website: https://yuchenlin.xyz/LLM-Blender/
☆ Learning to Relate to Previous Turns in Conversational Search KDD 2023
Conversational search allows a user to interact with a search system in multiple turns. A query is strongly dependent on the conversation context. An effective way to improve retrieval effectiveness is to expand the current query with historical queries. However, not all the previous queries are related to, and useful for expanding the current query. In this paper, we propose a new method to select relevant historical queries that are useful for the current query. To cope with the lack of labeled training data, we use a pseudo-labeling approach to annotate useful historical queries based on their impact on the retrieval results. The pseudo-labeled data are used to train a selection model. We further propose a multi-task learning framework to jointly train the selector and the retriever during fine-tuning, allowing us to mitigate the possible inconsistency between the pseudo labels and the changed retriever. Extensive experiments on four conversational search datasets demonstrate the effectiveness and broad applicability of our method compared with several strong baselines.
comment: Accepted by SIGKDD 2023 Research Track
☆ Evaluation of AI Chatbots for Patient-Specific EHR Questions
This paper investigates the use of artificial intelligence chatbots for patient-specific question answering (QA) from clinical notes using several large language model (LLM) based systems: ChatGPT (versions 3.5 and 4), Google Bard, and Claude. We evaluate the accuracy, relevance, comprehensiveness, and coherence of the answers generated by each model using a 5-point Likert scale on a set of patient-specific questions.
☆ Incorporating L2 Phonemes Using Articulatory Features for Robust Speech Recognition INTERSPEECH 2023
The limited availability of non-native speech datasets presents a major challenge in automatic speech recognition (ASR) to narrow the performance gap between native and non-native speakers. To address this, the focus of this study is on the efficient incorporation of the L2 phonemes, which in this work refer to Korean phonemes, through articulatory feature analysis. This not only enables accurate modeling of pronunciation variants but also allows for the utilization of both native Korean and English speech datasets. We employ the lattice-free maximum mutual information (LF-MMI) objective in an end-to-end manner, to train the acoustic model to align and predict one of multiple pronunciation candidates. Experimental results show that the proposed method improves ASR accuracy for Korean L2 speech by training solely on L1 speech data. Furthermore, fine-tuning on L2 speech improves recognition accuracy for both L1 and L2 speech without performance trade-offs.
comment: Accepted at INTERSPEECH 2023
☆ PLANNER: Generating Diversified Paragraph via Latent Language Diffusion Model
Autoregressive models for text sometimes generate repetitive and low-quality output because errors accumulate during the steps of generation. This issue is often attributed to exposure bias - the difference between how a model is trained, and how it is used during inference. Denoising diffusion models provide an alternative approach in which a model can revisit and revise its output. However, they can be computationally expensive and prior efforts on text have led to models that produce less fluent output compared to autoregressive models, especially for longer text and paragraphs. In this paper, we propose PLANNER, a model that combines latent semantic diffusion with autoregressive generation, to generate fluent text while exercising global control over paragraphs. The model achieves this by combining an autoregressive "decoding" module with a "planning" module that uses latent diffusion to generate semantic paragraph embeddings in a coarse-to-fine manner. The proposed method is evaluated on various conditional generation tasks, and results on semantic generation, text completion and summarization show its effectiveness in generating high-quality long-form text in an efficient manner.
☆ A Study of Situational Reasoning for Traffic Understanding KDD 2023
Intelligent Traffic Monitoring (ITMo) technologies hold the potential for improving road safety/security and for enabling smart city infrastructure. Understanding traffic situations requires a complex fusion of perceptual information with domain-specific and causal commonsense knowledge. Whereas prior work has provided benchmarks and methods for traffic monitoring, it remains unclear whether models can effectively align these information sources and reason in novel scenarios. To address this assessment gap, we devise three novel text-based tasks for situational reasoning in the traffic domain: i) BDD-QA, which evaluates the ability of Language Models (LMs) to perform situational decision-making, ii) TV-QA, which assesses LMs' abilities to reason about complex event causality, and iii) HDT-QA, which evaluates the ability of models to solve human driving exams. We adopt four knowledge-enhanced methods that have shown generalization capability across language reasoning tasks in prior work, based on natural language inference, commonsense knowledge-graph self-supervision, multi-QA joint training, and dense retrieval of domain information. We associate each method with a relevant knowledge source, including knowledge graphs, relevant benchmarks, and driving manuals. In extensive experiments, we benchmark various knowledge-aware methods against the three datasets, under zero-shot evaluation; we provide in-depth analyses of model performance on data partitions and examine model predictions categorically, to yield useful insights on traffic understanding, given different background knowledge and reasoning strategies.
comment: 11 pages, 6 figures, 5 tables, camera ready version of SIGKDD 2023
☆ Jambu: A historical linguistic database for South Asian languages
We introduce Jambu, a cognate database of South Asian languages which unifies dozens of previous sources in a structured and accessible format. The database includes 287k lemmata from 602 lects, grouped together in 23k sets of cognates. We outline the data wrangling necessary to compile the dataset and train neural models for reflex prediction on the Indo-Aryan subset of the data. We hope that Jambu is an invaluable resource for all historical linguists and Indologists, and look towards further improvement and expansion of the database.
comment: 5 pages main text, 10 pages total. To appear at SIGMORPHON
☆ CoSiNES: Contrastive Siamese Network for Entity Standardization ACL2023
Entity standardization maps noisy mentions from free-form text to standard entities in a knowledge base. The unique challenge of this task relative to other entity-related tasks is the lack of surrounding context and numerous variations in the surface form of the mentions, especially when it comes to generalization across domains where labeled data is scarce. Previous research mostly focuses on developing models either heavily relying on context, or dedicated solely to a specific domain. In contrast, we propose CoSiNES, a generic and adaptable framework with Contrastive Siamese Network for Entity Standardization that effectively adapts a pretrained language model to capture the syntax and semantics of the entities in a new domain. We construct a new dataset in the technology domain, which contains 640 technical stack entities and 6,412 mentions collected from industrial content management systems. We demonstrate that CoSiNES yields higher accuracy and faster runtime than baselines derived from leading methods in this domain. CoSiNES also achieves competitive performance in four standard datasets from the chemistry, medicine, and biomedical domains, demonstrating its cross-domain applicability.
comment: Accepted by Matching Workshop at ACL2023
☆ Few Shot Rationale Generation using Self-Training with Dual Teachers ACL
Self-rationalizing models that also generate a free-text explanation for their predicted labels are an important tool to build trustworthy AI applications. Since generating explanations for annotated labels is a laborious and costly pro cess, recent models rely on large pretrained language models (PLMs) as their backbone and few-shot learning. In this work we explore a self-training approach leveraging both labeled and unlabeled data to further improve few-shot models, under the assumption that neither human written rationales nor annotated task labels are available at scale. We introduce a novel dual-teacher learning framework, which learns two specialized teacher models for task prediction and rationalization using self-training and distills their knowledge into a multi-tasking student model that can jointly generate the task label and rationale. Furthermore, we formulate a new loss function, Masked Label Regularization (MLR) which promotes explanations to be strongly conditioned on predicted labels. Evaluation on three public datasets demonstrate that the proposed methods are effective in modeling task labels and generating faithful rationales.
comment: ACL Findings 2023
☆ A Scalable and Adaptive System to Infer the Industry Sectors of Companies: Prompt + Model Tuning of Generative Language Models IJCAI2023
The Private Equity (PE) firms operate investment funds by acquiring and managing companies to achieve a high return upon selling. Many PE funds are thematic, meaning investment professionals aim to identify trends by covering as many industry sectors as possible, and picking promising companies within these sectors. So, inferring sectors for companies is critical to the success of thematic PE funds. In this work, we standardize the sector framework and discuss the typical challenges; we then introduce our sector inference system addressing these challenges. Specifically, our system is built on a medium-sized generative language model, finetuned with a prompt + model tuning procedure. The deployed model demonstrates a superior performance than the common baselines. The system has been serving many PE professionals for over a year, showing great scalability to data volume and adaptability to any change in sector framework and/or annotation.
comment: Accepted by FinNLP (Financial Technology and Natural Language Processing) @ IJCAI2023 as long paper (8 pages and 8 figures)
☆ Stack Over-Flowing with Results: The Case for Domain-Specific Pre-Training Over One-Size-Fits-All Models
Large pre-trained neural language models have brought immense progress to both NLP and software engineering. Models in OpenAI's GPT series now dwarf Google's BERT and Meta's RoBERTa, which previously set new benchmarks on a wide range of NLP applications. These models are trained on massive corpora of heterogeneous data from web crawls, which enables them to learn general language patterns and semantic relationships. However, the largest models are both expensive to train and deploy and are often closed-source, so we lack access to their data and design decisions. We argue that this trend towards large, general-purpose models should be complemented with single-purpose, more modestly sized pre-trained models. In this work, we take StackOverflow (SO) as a domain example in which large volumes of rich aligned code and text data is available. We adopt standard practices for pre-training large language models, including using a very large context size (2,048 tokens), batch size (0.5M tokens) and training set (27B tokens), coupled with a powerful toolkit (Megatron-LM), to train two models: SOBertBase, with 109M parameters, and SOBertLarge with 762M parameters, at a budget of just $187 and $800 each. We compare the performance of our models with both the previous SOTA model trained on SO data exclusively as well general-purpose BERT models and OpenAI's ChatGPT on four SO-specific downstream tasks - question quality prediction, closed question prediction, named entity recognition and obsoletion prediction (a new task we introduce). Not only do our models consistently outperform all baselines, the smaller model is often sufficient for strong results. Both models are released to the public. These results demonstrate that pre-training both extensively and properly on in-domain data can yield a powerful and affordable alternative to leveraging closed-source general-purpose models.
☆ shs-nlp at RadSum23: Domain-Adaptive Pre-training of Instruction-tuned LLMs for Radiology Report Impression Generation
Instruction-tuned generative Large language models (LLMs) like ChatGPT and Bloomz possess excellent generalization abilities, but they face limitations in understanding radiology reports, particularly in the task of generating the IMPRESSIONS section from the FINDINGS section. They tend to generate either verbose or incomplete IMPRESSIONS, mainly due to insufficient exposure to medical text data during training. We present a system which leverages large-scale medical text data for domain-adaptive pre-training of instruction-tuned LLMs to enhance its medical knowledge and performance on specific medical tasks. We show that this system performs better in a zero-shot setting than a number of pretrain-and-finetune adaptation methods on the IMPRESSIONS generation task, and ranks 1st among participating systems in Task 1B: Radiology Report Summarization at the BioNLP 2023 workshop.
comment: 1st Place in Task 1B: Radiology Report Summarization at BioNLP 2023
☆ Understanding the Effectiveness of Early Weight Averaging for Training Large Language Models
Training LLMs is expensive, and recent evidence indicates training all the way to convergence is inefficient. In this paper, we investigate the ability of a simple idea, checkpoint averaging along the trajectory of a training run to improve the quality of models before they have converged. This approach incurs no extra cost during training or inference. Specifically, we analyze the training trajectories of Pythia LLMs with 1 to 12 billion parameters and demonstrate that, particularly during the early to mid stages of training, this idea accelerates convergence and improves both test and zero-shot generalization. Loss spikes are a well recognized problem in LLM training; in our analysis we encountered two instances of this in the underlying trajectories, and both instances were mitigated by our averaging. For a 6.9B parameter LLM, for example, our early weight averaging recipe can save upto 4200 hours of GPU time, which corresponds to significant savings in cloud compute costs.
comment: 17 pages, 12 figures, under review
☆ NLU on Data Diets: Dynamic Data Subset Selection for NLP Classification Tasks
Finetuning large language models inflates the costs of NLU applications and remains the bottleneck of development cycles. Recent works in computer vision use data pruning to reduce training time. Pruned data selection with static methods is based on a score calculated for each training example prior to finetuning, which involves important computational overhead. Moreover, the score may not necessarily be representative of sample importance throughout the entire training duration. We propose to address these issues with a refined version of dynamic data pruning, a curriculum which periodically scores and discards unimportant examples during finetuning. Our method leverages an EL2N metric that we extend to the joint intent and slot classification task, and an initial finetuning phase on the full train set. Our results on the GLUE benchmark and four joint NLU datasets show a better time-accuracy trade-off compared to static methods. Our method preserves full accuracy while training on 50% of the data points and reduces computational times by up to 41%. If we tolerate instead a minor drop of accuracy of 1%, we can prune 80% of the training examples for a reduction in finetuning time reaching 66%.
☆ A Static Evaluation of Code Completion by Large Language Models ACL 2023
Large language models trained on code have shown great potential to increase productivity of software developers. Several execution-based benchmarks have been proposed to evaluate functional correctness of model-generated code on simple programming problems. Nevertheless, it is expensive to perform the same evaluation on complex real-world projects considering the execution cost. On the contrary, static analysis tools such as linters, which can detect errors without running the program, haven't been well explored for evaluating code generation models. In this work, we propose a static evaluation framework to quantify static errors in Python code completions, by leveraging Abstract Syntax Trees. Compared with execution-based evaluation, our method is not only more efficient, but also applicable to code in the wild. For experiments, we collect code context from open source repos to generate one million function bodies using public models. Our static analysis reveals that Undefined Name and Unused Variable are the most common errors among others made by language models. Through extensive studies, we also show the impact of sampling temperature, model size, and context on static errors in code completions.
comment: Accepted by ACL 2023 industry track
☆ AutoScrum: Automating Project Planning Using Large Language Models
Recent advancements in the field of large language models have made it possible to use language models for advanced reasoning. In this paper we leverage this ability for designing complex project plans based only on knowing the current state and the desired state. Two approaches are demonstrated - a scrum based approach and a shortcut plan approach. The scrum based approach executes an automated process of requirements gathering, user story mapping, feature identification, task decomposition and finally generates questions and search terms for seeking out domain specific information to assist with task completion. The shortcut approach looks at most recent snapshot of the current and desired state and generates the next most reasonable task to do in order to get to the desired state as quickly as possible. In this paper we automate everything using a novel concept of "Language Programs". These are programs written in natural language designed to process input data through the language model. Guidance language is used for all LLM programs. All demo source code for this paper is available at https://github.com/autoscrum/autoscrum
comment: 25 pages, 3 figures, demo: https://github.com/autoscrum/autoscrum
☆ Easy-to-Read in Germany: A Survey on its Current State and Available Resources
Easy-to-Read Language (E2R) is a controlled language variant that makes any written text more accessible through the use of clear, direct and simple language. It is mainly aimed at people with cognitive or intellectual disabilities, among other target users. Plain Language (PL), on the other hand, is a variant of a given language, which aims to promote the use of simple language to communicate information. German counts with Leichte Sprache (LS), its version of E2R, and Einfache Sprache (ES), its version of PL. In recent years, important developments have been conducted in the field of LS. This paper offers an updated overview of the existing Natural Language Processing (NLP) tools and resources for LS. Besides, it also aims to set out the situation with regard to LS and ES in Germany.
comment: 10th Language & Technology Conference: Human Language Technologies as a Challenge for Computer Science and Linguistics, 2023
☆ Composition and Deformance: Measuring Imageability with a Text-to-Image Model
Although psycholinguists and psychologists have long studied the tendency of linguistic strings to evoke mental images in hearers or readers, most computational studies have applied this concept of imageability only to isolated words. Using recent developments in text-to-image generation models, such as DALLE mini, we propose computational methods that use generated images to measure the imageability of both single English words and connected text. We sample text prompts for image generation from three corpora: human-generated image captions, news article sentences, and poem lines. We subject these prompts to different deformances to examine the model's ability to detect changes in imageability caused by compositional change. We find high correlation between the proposed computational measures of imageability and human judgments of individual words. We also find the proposed measures more consistently respond to changes in compositionality than baseline approaches. We discuss possible effects of model training and implications for the study of compositionality in text-to-image models.
☆ Unsupervised Dense Retrieval with Relevance-Aware Contrastive Pre-Training ACL 2023
Dense retrievers have achieved impressive performance, but their demand for abundant training data limits their application scenarios. Contrastive pre-training, which constructs pseudo-positive examples from unlabeled data, has shown great potential to solve this problem. However, the pseudo-positive examples crafted by data augmentations can be irrelevant. To this end, we propose relevance-aware contrastive learning. It takes the intermediate-trained model itself as an imperfect oracle to estimate the relevance of positive pairs and adaptively weighs the contrastive loss of different pairs according to the estimated relevance. Our method consistently improves the SOTA unsupervised Contriever model on the BEIR and open-domain QA retrieval benchmarks. Further exploration shows that our method can not only beat BM25 after further pre-training on the target corpus but also serves as a good few-shot learner. Our code is publicly available at https://github.com/Yibin-Lei/ReContriever.
comment: ACL 2023 Findings (Short), 5 pages main + 1 page references + 1 page appendix
♻ ☆ PaLI: A Jointly-Scaled Multilingual Language-Image Model ICLR 2023
Effective scaling and a flexible task interface enable large language models to excel at many tasks. We present PaLI (Pathways Language and Image model), a model that extends this approach to the joint modeling of language and vision. PaLI generates text based on visual and textual inputs, and with this interface performs many vision, language, and multimodal tasks, in many languages. To train PaLI, we make use of large pre-trained encoder-decoder language models and Vision Transformers (ViTs). This allows us to capitalize on their existing capabilities and leverage the substantial cost of training them. We find that joint scaling of the vision and language components is important. Since existing Transformers for language are much larger than their vision counterparts, we train a large, 4-billion parameter ViT (ViT-e) to quantify the benefits from even larger-capacity vision models. To train PaLI, we create a large multilingual mix of pretraining tasks, based on a new image-text training set containing 10B images and texts in over 100 languages. PaLI achieves state-of-the-art in multiple vision and language tasks (such as captioning, visual question-answering, scene-text understanding), while retaining a simple, modular, and scalable design.
comment: ICLR 2023 (Notable-top-5%)
♻ ☆ In-context Example Selection with Influences
In-context learning (ICL) is a powerful paradigm emerged from large language models (LLMs). Despite its promises, ICL performance is known to be highly sensitive to input examples. In this work, we use $\textit{in-context influences}$ to analyze few-shot ICL performance directly from the in-context examples. Our proposed influence-based example selection method can identify both positive and negative examples, outperforming several baselines when evaluated on 9 SuperGLUE tasks. Our analysis uncovers up to a $16.3\%$ performance gap between using the most negative in-context examples compared to the most positive. In a case study, we apply our influence-based framework to quantify the phenomena of recency bias in example ordering for few-shot ICL.
♻ ☆ NusaCrowd: Open Source Initiative for Indonesian NLP Resources
We present NusaCrowd, a collaborative initiative to collect and unify existing resources for Indonesian languages, including opening access to previously non-public resources. Through this initiative, we have brought together 137 datasets and 118 standardized data loaders. The quality of the datasets has been assessed manually and automatically, and their value is demonstrated through multiple experiments. NusaCrowd's data collection enables the creation of the first zero-shot benchmarks for natural language understanding and generation in Indonesian and the local languages of Indonesia. Furthermore, NusaCrowd brings the creation of the first multilingual automatic speech recognition benchmark in Indonesian and the local languages of Indonesia. Our work strives to advance natural language processing (NLP) research for languages that are under-represented despite being widely spoken.
♻ ☆ A Systematic Study and Comprehensive Evaluation of ChatGPT on Benchmark Datasets ACL 2023
The development of large language models (LLMs) such as ChatGPT has brought a lot of attention recently. However, their evaluation in the benchmark academic datasets remains under-explored due to the difficulty of evaluating the generative outputs produced by this model against the ground truth. In this paper, we aim to present a thorough evaluation of ChatGPT's performance on diverse academic datasets, covering tasks like question-answering, text summarization, code generation, commonsense reasoning, mathematical problem-solving, machine translation, bias detection, and ethical considerations. Specifically, we evaluate ChatGPT across 140 tasks and analyze 255K responses it generates in these datasets. This makes our work the largest evaluation of ChatGPT in NLP benchmarks. In short, our study aims to validate the strengths and weaknesses of ChatGPT in various tasks and provide insights for future research using LLMs. We also report a new emergent ability to follow multi-query instructions that we mostly found in ChatGPT and other instruction-tuned models. Our extensive evaluation shows that even though ChatGPT is capable of performing a wide variety of tasks, and may obtain impressive performance in several benchmark datasets, it is still far from achieving the ability to reliably solve many challenging tasks. By providing a thorough assessment of ChatGPT's performance across diverse NLP tasks, this paper sets the stage for a targeted deployment of ChatGPT-like LLMs in real-world applications.
comment: Accepted by ACL 2023 Findings. The first three authors contributed equally
♻ ☆ Evaluating Inter-Bilingual Semantic Parsing for Indian Languages
Despite significant progress in Natural Language Generation for Indian languages (IndicNLP), there is a lack of datasets around complex structured tasks such as semantic parsing. One reason for this imminent gap is the complexity of the logical form, which makes English to multilingual translation difficult. The process involves alignment of logical forms, intents and slots with translated unstructured utterance. To address this, we propose an Inter-bilingual Seq2seq Semantic parsing dataset IE-SEMPARSE for 11 distinct Indian languages. We highlight the proposed task's practicality, and evaluate existing multilingual seq2seq models across several train-test strategies. Our experiment reveals a high correlation across performance of original multilingual semantic parsing datasets (such as mTOP, multilingual TOP and multiATIS++) and our proposed IE-SEMPARSE suite.
comment: 21 pages, 9 figures, 15 tables
♻ ☆ Tutel: Adaptive Mixture-of-Experts at Scale
Sparsely-gated mixture-of-experts (MoE) has been widely adopted to scale deep learning models to trillion-plus parameters with fixed computational cost. The algorithmic performance of MoE relies on its token routing mechanism that forwards each input token to the right sub-models or experts. While token routing dynamically determines the amount of expert workload at runtime, existing systems suffer inefficient computation due to their static execution, namely static parallelism and pipelining, which does not adapt to the dynamic workload. We present Flex, a highly scalable stack design and implementation for MoE with dynamically adaptive parallelism and pipelining. Flex designs an identical layout for distributing MoE model parameters and input data, which can be leveraged by all possible parallelism or pipelining methods without any mathematical inequivalence or tensor migration overhead. This enables adaptive parallelism/pipelining optimization at zero cost during runtime. Based on this key design, Flex also implements various MoE acceleration techniques. Aggregating all techniques, Flex finally delivers huge speedup at any scale -- 4.96x and 5.75x speedup of a single MoE layer over 16 and 2,048 A100 GPUs, respectively, over the previous state-of-the-art. Our evaluation shows that Flex efficiently and effectively runs a real-world MoE-based model named SwinV2-MoE, built upon Swin Transformer V2, a state-of-the-art computer vision architecture. On efficiency, Flex accelerates SwinV2-MoE, achieving up to 1.55x and 2.11x speedup in training and inference over Fairseq, respectively. On effectiveness, the SwinV2-MoE model achieves superior accuracy in both pre-training and down-stream computer vision tasks such as COCO object detection than the counterpart dense model, indicating the readiness of Flex for end-to-end real-world model training and inference.
♻ ☆ MAMO: Masked Multimodal Modeling for Fine-Grained Vision-Language Representation Learning SIGIR 2023
Multimodal representation learning has shown promising improvements on various vision-language tasks. Most existing methods excel at building global-level alignment between vision and language while lacking effective fine-grained image-text interaction. In this paper, we propose a jointly masked multimodal modeling method to learn fine-grained multimodal representations. Our method performs joint masking on image-text input and integrates both implicit and explicit targets for the masked signals to recover. The implicit target provides a unified and debiased objective for vision and language, where the model predicts latent multimodal representations of the unmasked input. The explicit target further enriches the multimodal representations by recovering high-level and semantically meaningful information: momentum visual features of image patches and concepts of word tokens. Through such a masked modeling process, our model not only learns fine-grained multimodal interaction, but also avoids the semantic gap between high-level representations and low- or mid-level prediction targets (e.g. image pixels), thus producing semantically rich multimodal representations that perform well on both zero-shot and fine-tuned settings. Our pre-trained model (named MAMO) achieves state-of-the-art performance on various downstream vision-language tasks, including image-text retrieval, visual question answering, visual reasoning, and weakly-supervised visual grounding.
comment: SIGIR 2023, 10 pages
♻ ☆ Signs of Language: Embodied Sign Language Fingerspelling Acquisition from Demonstrations for Human-Robot Interaction
Learning fine-grained movements is a challenging topic in robotics, particularly in the context of robotic hands. One specific instance of this challenge is the acquisition of fingerspelling sign language in robots. In this paper, we propose an approach for learning dexterous motor imitation from video examples without additional information. To achieve this, we first build a URDF model of a robotic hand with a single actuator for each joint. We then leverage pre-trained deep vision models to extract the 3D pose of the hand from RGB videos. Next, using state-of-the-art reinforcement learning algorithms for motion imitation (namely, proximal policy optimization and soft actor-critic), we train a policy to reproduce the movement extracted from the demonstrations. We identify the optimal set of hyperparameters for imitation based on a reference motion. Finally, we demonstrate the generalizability of our approach by testing it on six different tasks, corresponding to fingerspelled letters. Our results show that our approach is able to successfully imitate these fine-grained movements without additional information, highlighting its potential for real-world applications in robotics.
♻ ☆ Pre-training for Speech Translation: CTC Meets Optimal Transport ICML 2023
The gap between speech and text modalities is a major challenge in speech-to-text translation (ST). Different methods have been proposed to reduce this gap, but most of them require architectural changes in ST training. In this work, we propose to mitigate this issue at the pre-training stage, requiring no change in the ST model. First, we show that the connectionist temporal classification (CTC) loss can reduce the modality gap by design. We provide a quantitative comparison with the more common cross-entropy loss, showing that pre-training with CTC consistently achieves better final ST accuracy. Nevertheless, CTC is only a partial solution and thus, in our second contribution, we propose a novel pre-training method combining CTC and optimal transport to further reduce this gap. Our method pre-trains a Siamese-like model composed of two encoders, one for acoustic inputs and the other for textual inputs, such that they produce representations that are close to each other in the Wasserstein space. Extensive experiments on the standard CoVoST-2 and MuST-C datasets show that our pre-training method applied to the vanilla encoder-decoder Transformer achieves state-of-the-art performance under the no-external-data setting, and performs on par with recent strong multi-task learning systems trained with external data. Finally, our method can also be applied on top of these multi-task systems, leading to further improvements for these models. Code and pre-trained models are available at https://github.com/formiel/fairseq.
comment: ICML 2023 (oral presentation). This version fixed URLs, updated affiliations & acknowledgements, and improved formatting
♻ ☆ Boosting Radiology Report Generation by Infusing Comparison Prior ACL 2023
Recent transformer-based models have made significant strides in generating radiology reports from chest X-ray images. However, a prominent challenge remains: these models often lack prior knowledge, resulting in the generation of synthetic reports that mistakenly reference non-existent prior exams. This discrepancy can be attributed to a knowledge gap between radiologists and the generation models. While radiologists possess patient-specific prior information, the models solely receive X-ray images at a specific time point. To tackle this issue, we propose a novel approach that leverages a rule-based labeler to extract comparison prior information from radiology reports. This extracted comparison prior is then seamlessly integrated into state-of-the-art transformer-based models, enabling them to produce more realistic and comprehensive reports. Our method is evaluated on English report datasets, such as IU X-ray and MIMIC-CXR. The results demonstrate that our approach surpasses baseline models in terms of natural language generation metrics. Notably, our model generates reports that are free from false references to non-existent prior exams, setting it apart from previous models. By addressing this limitation, our approach represents a significant step towards bridging the gap between radiologists and generation models in the domain of medical report generation.
comment: Accepted at ACL 2023, BioNLP Workshop
♻ ☆ Language Anisotropic Cross-Lingual Model Editing
Multilingual pre-trained language models can learn task-specific abilities or memorize facts across multiple languages but inevitably make undesired predictions with specific inputs. Under similar observation, model editing aims to post-hoc calibrate a model targeted to specific inputs with keeping the model's raw behavior. However, existing work only studies the monolingual scenario, which lacks the cross-lingual transferability to perform editing simultaneously across languages. In this work, we focus on cross-lingual model editing. Firstly, we define the cross-lingual model editing task and corresponding metrics, where an edit in one language propagates to the others. Next, we propose a framework to naturally adapt monolingual model editing approaches to the cross-lingual scenario using parallel corpus. Further, we propose language anisotropic editing to improve cross-lingual editing by amplifying different subsets of parameters for each language. On the newly defined cross-lingual model editing task, we empirically demonstrate the failure of monolingual baselines in propagating the edit to multiple languages and the effectiveness of the proposed language anisotropic model editing. Our code is publicly available at https://github.com/franklear/LiME.
♻ ☆ Assessing the Importance of Frequency versus Compositionality for Subword-based Tokenization in NMT
Subword tokenization is the de facto standard for tokenization in neural language models and machine translation systems. Three advantages are frequently cited in favor of subwords: shorter encoding of frequent tokens, compositionality of subwords, and ability to deal with unknown words. As their relative importance is not entirely clear yet, we propose a tokenization approach that enables us to separate frequency (the first advantage) from compositionality. The approach uses Huffman coding to tokenize words, by order of frequency, using a fixed amount of symbols. Experiments with CS-DE, EN-FR and EN-DE NMT show that frequency alone accounts for 90%-95% of the scores reached by BPE, hence compositionality has less importance than previously thought.
comment: Accepted at EAMT 2023
♻ ☆ Evaluating Generative Patent Language Models
Generative language models are promising for assisting human writing in various domains. This manuscript aims to build generative language models in the patent domain and evaluate model performance from a human-centric perspective. The perspective is to measure the ratio of keystrokes that can be saved by autocompletion based on generative patent language models. A higher ratio means a more effective model which can save more keystrokes. This metric can be used to benchmark model performance. The metric is different from conventional machine-centric metrics that are token-based instead of keystroke-based. In terms of model size, the largest model built in this manuscript is 6B, which is state-of-the-art in the patent domain. Based on the metric, it is found that the largest model is not necessarily the best for the human-centric metric. The finding means that keeping increasing model sizes in the patent domain might be unnecessary if the purpose is to assist human writing with autocompletion. Several patent language models are pre-trained from scratch in this research. The pre-trained models are released for future researchers. Several visualization tools are also provided. The importance of building a generative language model in the patent domain is the potential to facilitate creativity and innovations in the future.
comment: 12 pages, 7 figures, and 5 tables
♻ ☆ A Unified Knowledge Graph Augmentation Service for Boosting Domain-specific NLP Tasks ACL
By focusing the pre-training process on domain-specific corpora, some domain-specific pre-trained language models (PLMs) have achieved state-of-the-art results. However, it is under-investigated to design a unified paradigm to inject domain knowledge in the PLM fine-tuning stage. We propose KnowledgeDA, a unified domain language model development service to enhance the task-specific training procedure with domain knowledge graphs. Given domain-specific task texts input, KnowledgeDA can automatically generate a domain-specific language model following three steps: (i) localize domain knowledge entities in texts via an embedding-similarity approach; (ii) generate augmented samples by retrieving replaceable domain entity pairs from two views of both knowledge graph and training data; (iii) select high-quality augmented samples for fine-tuning via confidence-based assessment. We implement a prototype of KnowledgeDA to learn language models for two domains, healthcare and software development. Experiments on domain-specific text classification and QA tasks verify the effectiveness and generalizability of KnowledgeDA.
comment: Accepted by ACL Findings 2023
♻ ☆ In-Context Analogical Reasoning with Pre-Trained Language Models
Analogical reasoning is a fundamental capacity of human cognition that allows us to reason abstractly about novel situations by relating them to past experiences. While it is thought to be essential for robust reasoning in AI systems, conventional approaches require significant training and/or hard-coding of domain knowledge to be applied to benchmark tasks. Inspired by cognitive science research that has found connections between human language and analogy-making, we explore the use of intuitive language-based abstractions to support analogy in AI systems. Specifically, we apply large pre-trained language models (PLMs) to visual Raven's Progressive Matrices (RPM), a common relational reasoning test. By simply encoding the perceptual features of the problem into language form, we find that PLMs exhibit a striking capacity for zero-shot relational reasoning, exceeding human performance and nearing supervised vision-based methods. We explore different encodings that vary the level of abstraction over task features, finding that higher-level abstractions further strengthen PLMs' analogical reasoning. Our detailed analysis reveals insights on the role of model complexity, in-context learning, and prior knowledge in solving RPM tasks.
♻ ☆ Self-Edit: Fault-Aware Code Editor for Code Generation ACL2023
Large language models (LLMs) have demonstrated an impressive ability to generate codes on competitive programming tasks. However, with limited sample numbers, LLMs still suffer from poor accuracy. Inspired by the process of human programming, we propose a generate-and-edit approach named Self-Edit that utilizes execution results of the generated code from LLMs to improve the code quality on the competitive programming task. We execute the generated code on the example test case provided in the question and wrap execution results into a supplementary comment. Utilizing this comment as guidance, our fault-aware code editor is employed to correct errors in the generated code. We perform extensive evaluations across two competitive programming datasets with nine different LLMs. Compared to directly generating from LLMs, our approach can improve the average of pass@1 by 89\% on APPS-dev, 31\% on APPS-test, and 48\% on HumanEval over nine popular code generation LLMs with parameter sizes ranging from 110M to 175B. Compared to other post-processing methods, our method demonstrates superior accuracy and efficiency.
comment: Accepted by ACL2023
♻ ☆ Dataless Knowledge Fusion by Merging Weights of Language Models ICLR 2023
Fine-tuning pre-trained language models has become the prevalent paradigm for building downstream NLP models. Oftentimes fine-tuned models are readily available but their training data is not, due to data privacy or intellectual property concerns. This creates a barrier to fusing knowledge across individual models to yield a better single model. In this paper, we study the problem of merging individual models built on different training data sets to obtain a single model that performs well both across all data set domains and can generalize on out-of-domain data. We propose a dataless knowledge fusion method that merges models in their parameter space, guided by weights that minimize prediction differences between the merged model and the individual models. Over a battery of evaluation settings, we show that the proposed method significantly outperforms baselines such as Fisher-weighted averaging or model ensembling. Further, we find that our method is a promising alternative to multi-task learning that can preserve or sometimes improve over the individual models without access to the training data. Finally, model merging is more efficient than training a multi-task model, thus making it applicable to a wider set of scenarios.
comment: ICLR 2023; Updated captions of Table 1. The code is available at https://github.com/bloomberg/dataless-model-merging
♻ ☆ Large Language Models can be Guided to Evade AI-Generated Text Detection
Large Language Models (LLMs) have demonstrated exceptional performance in a variety of tasks, including essay writing and question answering. However, it is crucial to address the potential misuse of these models, which can lead to detrimental outcomes such as plagiarism and spamming. Recently, several detectors have been proposed, including fine-tuned classifiers and various statistical methods. In this study, we reveal that with the aid of carefully crafted prompts, LLMs can effectively evade these detection systems. We propose a novel Substitution-based In-Context example Optimization method (SICO) to automatically generate such prompts. On three real-world tasks where LLMs can be misused, SICO successfully enables ChatGPT to evade six existing detectors, causing a significant 0.54 AUC drop on average. Surprisingly, in most cases these detectors perform even worse than random classifiers. These results firmly reveal the vulnerability of existing detectors. Finally, the strong performance of SICO suggests itself as a reliable evaluation protocol for any new detector in this field.
♻ ☆ MGR: Multi-generator Based Rationalization ACL 2023
Rationalization is to employ a generator and a predictor to construct a self-explaining NLP model in which the generator selects a subset of human-intelligible pieces of the input text to the following predictor. However, rationalization suffers from two key challenges, i.e., spurious correlation and degeneration, where the predictor overfits the spurious or meaningless pieces solely selected by the not-yet well-trained generator and in turn deteriorates the generator. Although many studies have been proposed to address the two challenges, they are usually designed separately and do not take both of them into account. In this paper, we propose a simple yet effective method named MGR to simultaneously solve the two problems. The key idea of MGR is to employ multiple generators such that the occurrence stability of real pieces is improved and more meaningful pieces are delivered to the predictor. Empirically, we show that MGR improves the F1 score by up to 20.9% as compared to state-of-the-art methods. Codes are available at https://github.com/jugechengzi/Rationalization-MGR .
comment: Accepted as a main conference paper of ACL 2023. arXiv admin note: text overlap with arXiv:2209.08285
♻ ☆ BUMP: A Benchmark of Unfaithful Minimal Pairs for Meta-Evaluation of Faithfulness Metrics ACL 2023
The proliferation of automatic faithfulness metrics for summarization has produced a need for benchmarks to evaluate them. While existing benchmarks measure the correlation with human judgements of faithfulness on model-generated summaries, they are insufficient for diagnosing whether metrics are: 1) consistent, i.e., indicate lower faithfulness as errors are introduced into a summary, 2) effective on human-written texts, and 3) sensitive to different error types (as summaries can contain multiple errors). To address these needs, we present a benchmark of unfaithful minimal pairs (BUMP), a dataset of 889 human-written, minimally different summary pairs, where a single error is introduced to a summary from the CNN/DailyMail dataset to produce an unfaithful summary. We find BUMP complements existing benchmarks in a number of ways: 1) the summaries in BUMP are harder to discriminate and less probable under SOTA summarization models, 2) unlike non-pair-based datasets, BUMP can be used to measure the consistency of metrics, and reveals that the most discriminative metrics tend not to be the most consistent, and 3) unlike datasets containing generated summaries with multiple errors, BUMP enables the measurement of metrics' performance on individual error types.
comment: Accepted as a long main conference paper at ACL 2023
♻ ☆ Few-shot Domain-Adaptive Visually-fused Event Detection from Text
Incorporating auxiliary modalities such as images into event detection models has attracted increasing interest over the last few years. The complexity of natural language in describing situations has motivated researchers to leverage the related visual context to improve event detection performance. However, current approaches in this area suffer from data scarcity, where a large amount of labelled text-image pairs are required for model training. Furthermore, limited access to the visual context at inference time negatively impacts the performance of such models, which makes them practically ineffective in real-world scenarios. In this paper, we present a novel domain-adaptive visually-fused event detection approach that can be trained on a few labelled image-text paired data points. Specifically, we introduce a visual imaginator method that synthesises images from text in the absence of visual context. Moreover, the imaginator can be customised to a specific domain. In doing so, our model can leverage the capabilities of pre-trained vision-language models and can be trained in a few-shot setting. This also allows for effective inference where only single-modality data (i.e. text) is available. The experimental evaluation on the benchmark M2E2 dataset shows that our model outperforms existing state-of-the-art models, by up to 11 points.
♻ ☆ APOLLO: A Simple Approach for Adaptive Pretraining of Language Models for Logical Reasoning ACL 2023
Logical reasoning of text is an important ability that requires understanding the information present in the text, their interconnections, and then reasoning through them to infer new conclusions. Prior works on improving the logical reasoning ability of language models require complex processing of training data (e.g., aligning symbolic knowledge to text), yielding task-specific data augmentation solutions that restrict the learning of general logical reasoning skills. In this work, we propose APOLLO, an adaptively pretrained language model that has improved logical reasoning abilities. We select a subset of Wikipedia, based on a set of logical inference keywords, for continued pretraining of a language model. We use two self-supervised loss functions: a modified masked language modeling loss where only specific parts-of-speech words, that would likely require more reasoning than basic language understanding, are masked, and a sentence-level classification loss that teaches the model to distinguish between entailment and contradiction types of sentences. The proposed training paradigm is both simple and independent of task formats. We demonstrate the effectiveness of APOLLO by comparing it with prior baselines on two logical reasoning datasets. APOLLO performs comparably on ReClor and outperforms baselines on LogiQA. The code base has been made publicly available.
comment: Accepted at ACL 2023, code available at https://github.com/INK-USC/APOLLO
♻ ☆ Multi-hop Evidence Retrieval for Cross-document Relation Extraction ACL 2023
Relation Extraction (RE) has been extended to cross-document scenarios because many relations are not simply described in a single document. This inevitably brings the challenge of efficient open-space evidence retrieval to support the inference of cross-document relations, along with the challenge of multi-hop reasoning on top of entities and evidence scattered in an open set of documents. To combat these challenges, we propose MR.COD (Multi-hop evidence retrieval for Cross-document relation extraction), which is a multi-hop evidence retrieval method based on evidence path mining and ranking. We explore multiple variants of retrievers to show evidence retrieval is essential in cross-document RE. We also propose a contextual dense retriever for this setting. Experiments on CodRED show that evidence retrieval with MR.COD effectively acquires crossdocument evidence and boosts end-to-end RE performance in both closed and open settings.
comment: ACL 2023 (Findings)
♻ ☆ A Multilingual Perspective Towards the Evaluation of Attribution Methods in Natural Language Inference EMNLP 2022
Most evaluations of attribution methods focus on the English language. In this work, we present a multilingual approach for evaluating attribution methods for the Natural Language Inference (NLI) task in terms of faithfulness and plausibility. First, we introduce a novel cross-lingual strategy to measure faithfulness based on word alignments, which eliminates the drawbacks of erasure-based evaluations.We then perform a comprehensive evaluation of attribution methods, considering different output mechanisms and aggregation methods. Finally, we augment the XNLI dataset with highlight-based explanations, providing a multilingual NLI dataset with highlights, to support future exNLP studies. Our results show that attribution methods performing best for plausibility and faithfulness are different.
comment: 21 pages, 7 figures. Code and data at https://keremzaman.com/explaiNLI/; Published in the Proceedings of EMNLP 2022
♻ ☆ NarrowBERT: Accelerating Masked Language Model Pretraining and Inference ACL 2023
Large-scale language model pretraining is a very successful form of self-supervised learning in natural language processing, but it is increasingly expensive to perform as the models and pretraining corpora have become larger over time. We propose NarrowBERT, a modified transformer encoder that increases the throughput for masked language model pretraining by more than $2\times$. NarrowBERT sparsifies the transformer model such that the self-attention queries and feedforward layers only operate on the masked tokens of each sentence during pretraining, rather than all of the tokens as with the usual transformer encoder. We also show that NarrowBERT increases the throughput at inference time by as much as $3.5\times$ with minimal (or no) performance degradation on sentence encoding tasks like MNLI. Finally, we examine the performance of NarrowBERT on the IMDB and Amazon reviews classification and CoNLL NER tasks and show that it is also comparable to standard BERT performance.
comment: To appear in ACL 2023 (main conference)
♻ ☆ Less is More: Task-aware Layer-wise Distillation for Language Model Compression ICML 2023
Layer-wise distillation is a powerful tool to compress large models (i.e. teacher models) into small ones (i.e., student models). The student distills knowledge from the teacher by mimicking the hidden representations of the teacher at every intermediate layer. However, layer-wise distillation is difficult. Since the student has a smaller model capacity than the teacher, it is often under-fitted. Furthermore, the hidden representations of the teacher contain redundant information that the student does not necessarily need for the target task's learning. To address these challenges, we propose a novel Task-aware layEr-wise Distillation (TED). TED designs task-aware filters to align the hidden representations of the student and the teacher at each layer. The filters select the knowledge that is useful for the target task from the hidden representations. As such, TED reduces the knowledge gap between the two models and helps the student to fit better on the target task. We evaluate TED in two scenarios: continual pre-training and fine-tuning. TED demonstrates significant and consistent improvements over existing distillation methods in both scenarios. Code is available at https://github.com/cliang1453/task-aware-distillation.
comment: Proceedings of ICML 2023
♻ ☆ SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models ICML 2023
Large language models (LLMs) show excellent performance but are compute- and memory-intensive. Quantization can reduce memory and accelerate inference. However, existing methods cannot maintain accuracy and hardware efficiency at the same time. We propose SmoothQuant, a training-free, accuracy-preserving, and general-purpose post-training quantization (PTQ) solution to enable 8-bit weight, 8-bit activation (W8A8) quantization for LLMs. Based on the fact that weights are easy to quantize while activations are not, SmoothQuant smooths the activation outliers by offline migrating the quantization difficulty from activations to weights with a mathematically equivalent transformation. SmoothQuant enables an INT8 quantization of both weights and activations for all the matrix multiplications in LLMs, including OPT, BLOOM, GLM, MT-NLG, and LLaMA family. We demonstrate up to 1.56x speedup and 2x memory reduction for LLMs with negligible loss in accuracy. SmoothQuant enables serving 530B LLM within a single node. Our work offers a turn-key solution that reduces hardware costs and democratizes LLMs. Code is available at https://github.com/mit-han-lab/smoothquant.
comment: ICML 2023. First two authors contributed equally to this work
♻ ☆ Structured Knowledge Grounding for Question Answering
Can language models (LM) ground question-answering (QA) tasks in the knowledge base via inherent relational reasoning ability? While previous models that use only LMs have seen some success on many QA tasks, more recent methods include knowledge graphs (KG) to complement LMs with their more logic-driven implicit knowledge. However, effectively extracting information from structured data, like KGs, empowers LMs to remain an open question, and current models rely on graph techniques to extract knowledge. In this paper, we propose to solely leverage the LMs to combine the language and knowledge for knowledge based question-answering with flexibility, breadth of coverage and structured reasoning. Specifically, we devise a knowledge construction method that retrieves the relevant context with a dynamic hop, which expresses more comprehensivenes than traditional GNN-based techniques. And we devise a deep fusion mechanism to further bridge the information exchanging bottleneck between the language and the knowledge. Extensive experiments show that our model consistently demonstrates its state-of-the-art performance over CommensenseQA benchmark, showcasing the possibility to leverage LMs solely to robustly ground QA into the knowledge base.
♻ ☆ DISCO: Distilling Counterfactuals with Large Language Models ACL 2023
Models trained with counterfactually augmented data learn representations of the causal structure of tasks, enabling robust generalization. However, high-quality counterfactual data is scarce for most tasks and not easily generated at scale. When crowdsourced, such data is typically limited in scale and diversity; when generated using supervised methods, it is computationally expensive to extend to new counterfactual dimensions. In this work, we introduce DISCO (DIStilled COunterfactual Data), a new method for automatically generating high quality counterfactual data at scale. DISCO engineers prompts to generate phrasal perturbations with a large general language model. Then, a task-specific teacher model filters these generations to distill high-quality counterfactual data. While task-agnostic, we apply our pipeline to the task of natural language inference (NLI) and find that on challenging evaluations such as the NLI stress test, comparatively smaller student models trained with DISCO generated counterfactuals are more robust (6% absolute) and generalize better across distributions (2%) compared to models trained without data augmentation. Furthermore, DISCO augmented models are 10% more consistent between counterfactual pairs on three evaluation sets, demonstrating that DISCO augmentation enables models to more reliably learn causal representations. Our repository is available at: https://github.com/eric11eca/disco
comment: ACL 2023 camera ready, final title change
♻ ☆ Joint Repetition Suppression and Content Moderation of Large Language Models
Natural language generation (NLG) is one of the most impactful fields in NLP, and recent years have witnessed its evolution brought about by large language models (LLMs). As the key instrument for writing assistance applications, they are generally prone to replicating or extending offensive content provided in the input. In low-resource data regime, they can also lead to repetitive outputs. Usually, offensive content and repetitions are mitigated with post-hoc methods, including n-gram level blocklists, top-k and nucleus sampling. In this paper, we apply non-exact repetition suppression using token and sequence level unlikelihood loss, and further explore the framework of unlikelihood training objective in order to jointly endow the model with abilities to avoid generating offensive words and phrases from the beginning. Finally, with comprehensive experiments, we demonstrate that our proposed methods work exceptionally in controlling the repetition and content quality of LLM outputs.
Computer Vision and Pattern Recognition 126
☆ Neuralangelo: High-Fidelity Neural Surface Reconstruction CVPR 2023
Neural surface reconstruction has been shown to be powerful for recovering dense 3D surfaces via image-based neural rendering. However, current methods struggle to recover detailed structures of real-world scenes. To address the issue, we present Neuralangelo, which combines the representation power of multi-resolution 3D hash grids with neural surface rendering. Two key ingredients enable our approach: (1) numerical gradients for computing higher-order derivatives as a smoothing operation and (2) coarse-to-fine optimization on the hash grids controlling different levels of details. Even without auxiliary inputs such as depth, Neuralangelo can effectively recover dense 3D surface structures from multi-view images with fidelity significantly surpassing previous methods, enabling detailed large-scale scene reconstruction from RGB video captures.
comment: CVPR 2023, project page: https://research.nvidia.com/labs/dir/neuralangelo
☆ Brain Diffusion for Visual Exploration: Cortical Discovery using Large Scale Generative Models
A long standing goal in neuroscience has been to elucidate the functional organization of the brain. Within higher visual cortex, functional accounts have remained relatively coarse, focusing on regions of interest (ROIs) and taking the form of selectivity for broad categories such as faces, places, bodies, food, or words. Because the identification of such ROIs has typically relied on manually assembled stimulus sets consisting of isolated objects in non-ecological contexts, exploring functional organization without robust a priori hypotheses has been challenging. To overcome these limitations, we introduce a data-driven approach in which we synthesize images predicted to activate a given brain region using paired natural images and fMRI recordings, bypassing the need for category-specific stimuli. Our approach -- Brain Diffusion for Visual Exploration ("BrainDiVE") -- builds on recent generative methods by combining large-scale diffusion models with brain-guided image synthesis. Validating our method, we demonstrate the ability to synthesize preferred images with appropriate semantic specificity for well-characterized category-selective ROIs. We then show that BrainDiVE can characterize differences between ROIs selective for the same high-level category. Finally we identify novel functional subdivisions within these ROIs, validated with behavioral data. These results advance our understanding of the fine-grained functional organization of human visual cortex, and provide well-specified constraints for further examination of cortical organization using hypothesis-driven methods.
☆ Of Mice and Mates: Automated Classification and Modelling of Mouse Behaviour in Groups using a Single Model across Cages
Behavioural experiments often happen in specialised arenas, but this may confound the analysis. To address this issue, we provide tools to study mice in the homecage environment, equipping biologists with the possibility to capture the temporal aspect of the individual's behaviour and model the interaction and interdependence between cage-mates with minimal human intervention. We develop the Activity Labelling Module (ALM) to automatically classify mouse behaviour from video, and a novel Group Behaviour Model (GBM) for summarising their joint behaviour across cages, using a permutation matrix to match the mouse identities in each cage to the model. We also release two datasets, ABODe for training behaviour classifiers and IMADGE for modelling behaviour.
☆ ELEV-VISION: Automated Lowest Floor Elevation Estimation from Segmenting Street View Images
We propose an automated lowest floor elevation (LFE) estimation algorithm based on computer vision techniques to leverage the latent information in street view images. Flood depth-damage models use a combination of LFE and flood depth for determining flood risk and extent of damage to properties. We used image segmentation for detecting door bottoms and roadside edges from Google Street View images. The characteristic of equirectangular projection with constant spacing representation of horizontal and vertical angles allows extraction of the pitch angle from the camera to the door bottom. The depth from the camera to the door bottom was obtained from the depthmap paired with the Google Street View image. LFEs were calculated from the pitch angle and the depth. The testbed for application of the proposed method is Meyerland (Harris County, Texas). The results show that the proposed method achieved mean absolute error of 0.190 m (1.18 %) in estimating LFE. The height difference between the street and the lowest floor (HDSL) was estimated to provide information for flood damage estimation. The proposed automatic LFE estimation algorithm using Street View images and image segmentation provides a rapid and cost-effective method for LFE estimation compared with the surveys using total station theodolite and unmanned aerial systems. By obtaining more accurate and up-to-date LFE data using the proposed method, city planners, emergency planners and insurance companies could make a more precise estimation of flood damage.
☆ HeadSculpt: Crafting 3D Head Avatars with Text
Recently, text-guided 3D generative methods have made remarkable advancements in producing high-quality textures and geometry, capitalizing on the proliferation of large vision-language and image diffusion models. However, existing methods still struggle to create high-fidelity 3D head avatars in two aspects: (1) They rely mostly on a pre-trained text-to-image diffusion model whilst missing the necessary 3D awareness and head priors. This makes them prone to inconsistency and geometric distortions in the generated avatars. (2) They fall short in fine-grained editing. This is primarily due to the inherited limitations from the pre-trained 2D image diffusion models, which become more pronounced when it comes to 3D head avatars. In this work, we address these challenges by introducing a versatile coarse-to-fine pipeline dubbed HeadSculpt for crafting (i.e., generating and editing) 3D head avatars from textual prompts. Specifically, we first equip the diffusion model with 3D awareness by leveraging landmark-based control and a learned textual embedding representing the back view appearance of heads, enabling 3D-consistent head avatar generations. We further propose a novel identity-aware editing score distillation strategy to optimize a textured mesh with a high-resolution differentiable rendering technique. This enables identity preservation while following the editing instruction. We showcase HeadSculpt's superior fidelity and editing capabilities through comprehensive experiments and comparisons with existing methods.
comment: Webpage: https://brandonhan.uk/HeadSculpt/
☆ Interpretable Alzheimer's Disease Classification Via a Contrastive Diffusion Autoencoder
In visual object classification, humans often justify their choices by comparing objects to prototypical examples within that class. We may therefore increase the interpretability of deep learning models by imbuing them with a similar style of reasoning. In this work, we apply this principle by classifying Alzheimer's Disease based on the similarity of images to training examples within the latent space. We use a contrastive loss combined with a diffusion autoencoder backbone, to produce a semantically meaningful latent space, such that neighbouring latents have similar image-level features. We achieve a classification accuracy comparable to black box approaches on a dataset of 2D MRI images, whilst producing human interpretable model explanations. Therefore, this work stands as a contribution to the pertinent development of accurate and interpretable deep learning within medical imaging.
☆ Automating Style Analysis and Visualization With Explainable AI -- Case Studies on Brand Recognition
Incorporating style-related objectives into shape design has been centrally important to maximize product appeal. However, stylistic features such as aesthetics and semantic attributes are hard to codify even for experts. As such, algorithmic style capture and reuse have not fully benefited from automated data-driven methodologies due to the challenging nature of design describability. This paper proposes an AI-driven method to fully automate the discovery of brand-related features. Our approach introduces BIGNet, a two-tier Brand Identification Graph Neural Network (GNN) to classify and analyze scalar vector graphics (SVG). First, to tackle the scarcity of vectorized product images, this research proposes two data acquisition workflows: parametric modeling from small curve-based datasets, and vectorization from large pixel-based datasets. Secondly, this study constructs a novel hierarchical GNN architecture to learn from both SVG's curve-level and chunk-level parameters. In the first case study, BIGNet not only classifies phone brands but also captures brand-related features across multiple scales, such as the location of the lens, the height-width ratio, and the screen-frame gap, as confirmed by AI evaluation. In the second study, this paper showcases the generalizability of BIGNet learning from a vectorized car image dataset and validates the consistency and robustness of its predictions given four scenarios. The results match the difference commonly observed in luxury vs. economy brands in the automobile market. Finally, this paper also visualizes the activation maps generated from a convolutional neural network and shows BIGNet's advantage of being a more human-friendly, explainable, and explicit style-capturing agent. Code and dataset can be found on Github: 1. Phone case study: github.com/parksandrecfan/bignet-phone 2. Car case study: github.com/parksandrecfan/bignet-car
☆ Nonparametric Iterative Machine Teaching ICML 2023
In this paper, we consider the problem of Iterative Machine Teaching (IMT), where the teacher provides examples to the learner iteratively such that the learner can achieve fast convergence to a target model. However, existing IMT algorithms are solely based on parameterized families of target models. They mainly focus on convergence in the parameter space, resulting in difficulty when the target models are defined to be functions without dependency on parameters. To address such a limitation, we study a more general task -- Nonparametric Iterative Machine Teaching (NIMT), which aims to teach nonparametric target models to learners in an iterative fashion. Unlike parametric IMT that merely operates in the parameter space, we cast NIMT as a functional optimization problem in the function space. To solve it, we propose both random and greedy functional teaching algorithms. We obtain the iterative teaching dimension (ITD) of the random teaching algorithm under proper assumptions, which serves as a uniform upper bound of ITD in NIMT. Further, the greedy teaching algorithm has a significantly lower ITD, which reaches a tighter upper bound of ITD in NIMT. Finally, we verify the correctness of our theoretical findings with extensive experiments in nonparametric scenarios.
comment: ICML 2023 (20 pages, 10 figures)
☆ Unveiling the Two-Faced Truth: Disentangling Morphed Identities for Face Morphing Detection
Morphing attacks keep threatening biometric systems, especially face recognition systems. Over time they have become simpler to perform and more realistic, as such, the usage of deep learning systems to detect these attacks has grown. At the same time, there is a constant concern regarding the lack of interpretability of deep learning models. Balancing performance and interpretability has been a difficult task for scientists. However, by leveraging domain information and proving some constraints, we have been able to develop IDistill, an interpretable method with state-of-the-art performance that provides information on both the identity separation on morph samples and their contribution to the final prediction. The domain information is learnt by an autoencoder and distilled to a classifier system in order to teach it to separate identity information. When compared to other methods in the literature it outperforms them in three out of five databases and is competitive in the remaining.
comment: Accepted at EUSIPCO 2023
☆ BeyondPixels: A Comprehensive Review of the Evolution of Neural Radiance Fields
Neural rendering combines ideas from classical computer graphics and machine learning to synthesize images from real-world observations. NeRF, short for Neural Radiance Fields, is a recent innovation that uses AI algorithms to create 3D objects from 2D images. By leveraging an interpolation approach, NeRF can produce new 3D reconstructed views of complicated scenes. Rather than directly restoring the whole 3D scene geometry, NeRF generates a volumetric representation called a ``radiance field,'' which is capable of creating color and density for every point within the relevant 3D space. The broad appeal and notoriety of NeRF make it imperative to examine the existing research on the topic comprehensively. While previous surveys on 3D rendering have primarily focused on traditional computer vision-based or deep learning-based approaches, only a handful of them discuss the potential of NeRF. However, such surveys have predominantly focused on NeRF's early contributions and have not explored its full potential. NeRF is a relatively new technique continuously being investigated for its capabilities and limitations. This survey reviews recent advances in NeRF and categorizes them according to their architectural designs, especially in the field of novel view synthesis.
comment: 22 page, 1 figure, 1 table
☆ Long-range UAV Thermal Geo-localization with Satellite Imagery
Onboard sensors, such as cameras and thermal sensors, have emerged as effective alternatives to Global Positioning System (GPS) for geo-localization in Unmanned Aerial Vehicle (UAV) navigation. Since GPS can suffer from signal loss and spoofing problems, researchers have explored camera-based techniques such as Visual Geo-localization (VG) using satellite imagery. Additionally, thermal geo-localization (TG) has become crucial for long-range UAV flights in low-illumination environments. This paper proposes a novel thermal geo-localization framework using satellite imagery, which includes multiple domain adaptation methods to address the limited availability of paired thermal and satellite images. The experimental results demonstrate the effectiveness of the proposed approach in achieving reliable thermal geo-localization performance, even in thermal images with indistinct self-similar features. We evaluate our approach on real data collected onboard a UAV. We also release the code and \textit{Boson-nighttime}, a dataset of paired satellite-thermal and unpaired satellite images for thermal geo-localization with satellite imagery. To the best of our knowledge, this work is the first to propose a thermal geo-localization method using satellite imagery in long-range flights.
comment: 8 pages, 6 figures
☆ Brain tumor segmentation using synthetic MR images -- A comparison of GANs and diffusion models
Large annotated datasets are required for training deep learning models, but in medical imaging data sharing is often complicated due to ethics, anonymization and data protection legislation (e.g. the general data protection regulation (GDPR)). Generative AI models, such as generative adversarial networks (GANs) and diffusion models, can today produce very realistic synthetic images, and can potentially facilitate data sharing as GDPR should not apply for medical images which do not belong to a specific person. However, in order to share synthetic images it must first be demonstrated that they can be used for training different networks with acceptable performance. Here, we therefore comprehensively evaluate four GANs (progressive GAN, StyleGAN 1-3) and a diffusion model for the task of brain tumor segmentation. Our results show that segmentation networks trained on synthetic images reach Dice scores that are 80\% - 90\% of Dice scores when training with real images, but that memorization of the training images can be a problem for diffusion models if the original dataset is too small. Furthermore, we demonstrate that common metrics for evaluating synthetic images, Fr\'echet inception distance (FID) and inception score (IS), do not correlate well with the obtained performance when using the synthetic images for training segmentation networks.
comment: 20 Pages
☆ Best of Both Worlds: Hybrid SNN-ANN Architecture for Event-based Optical Flow Estimation
Event-based cameras offer a low-power alternative to frame-based cameras for capturing high-speed motion and high dynamic range scenes. They provide asynchronous streams of sparse events. Spiking Neural Networks (SNNs) with their asynchronous event-driven compute, show great potential for extracting the spatio-temporal features from these event streams. In contrast, the standard Analog Neural Networks (ANNs1) fail to process event data effectively. However, training SNNs is difficult due to additional trainable parameters (thresholds and leaks), vanishing spikes at deeper layers, non-differentiable binary activation function etc. Moreover, an additional data structure "membrane potential" responsible for keeping track of temporal information, must be fetched and updated at every timestep in SNNs. To overcome these, we propose a novel SNN-ANN hybrid architecture that combines the strengths of both. Specifically, we leverage the asynchronous compute capabilities of SNN layers to effectively extract the input temporal information. While the ANN layers offer trouble-free training and implementation on standard machine learning hardware such as GPUs. We provide extensive experimental analysis for assigning each layer to be spiking or analog in nature, leading to a network configuration optimized for performance and ease of training. We evaluate our hybrid architectures for optical flow estimation using event-data on DSEC-flow and Mutli-Vehicle Stereo Event-Camera (MVSEC) datasets. The results indicate that our configured hybrid architectures outperform the state-of-the-art ANN-only, SNN-only and past hybrid architectures both in terms of accuracy and efficiency. Specifically, our hybrid architecture exhibit a 31% and 24.8% lower average endpoint error (AEE) at 2.1x and 3.1x lower energy, compared to an SNN-only architecture on DSEC and MVSEC datasets, respectively.
☆ Explicit Neural Surfaces: Learning Continuous Geometry With Deformation Fields
We introduce Explicit Neural Surfaces (ENS), an efficient surface reconstruction method that learns an explicitly defined continuous surface from multiple views. We use a series of neural deformation fields to progressively transform a continuous input surface to a target shape. By sampling meshes as discrete surface proxies, we train the deformation fields through efficient differentiable rasterization, and attain a mesh-independent and smooth surface representation. By using Laplace-Beltrami eigenfunctions as an intrinsic positional encoding alongside standard extrinsic Fourier features, our approach can capture fine surface details. ENS trains 1 to 2 orders of magnitude faster and can extract meshes of higher quality compared to implicit representations, whilst maintaining competitive surface reconstruction performance and real-time capabilities. Finally, we apply our approach to learn a collection of objects in a single model, and achieve disentangled interpolations between different shapes, their surface details, and textures.
☆ Color-aware Deep Temporal Backdrop Duplex Matting System
Deep learning-based alpha matting showed tremendous improvements in recent years, yet, feature film production studios still rely on classical chroma keying including costly post-production steps. This perceived discrepancy can be explained by some missing links necessary for production which are currently not adequately addressed in the alpha matting community, in particular foreground color estimation or color spill compensation. We propose a neural network-based temporal multi-backdrop production system that combines beneficial features from chroma keying and alpha matting. Given two consecutive frames with different background colors, our one-encoder-dual-decoder network predicts foreground colors and alpha values using a patch-based overlap-blend approach. The system is able to handle imprecise backdrops, dynamic cameras, and dynamic foregrounds and has no restrictions on foreground colors. We compare our method to state-of-the-art algorithms using benchmark datasets and a video sequence captured by a demonstrator setup. We verify that a dual backdrop input is superior to the usually applied trimap-based approach. In addition, the proposed studio set is actor friendly, and produces high-quality, temporal consistent alpha and color estimations that include a superior color spill compensation.
☆ INDigo: An INN-Guided Probabilistic Diffusion Algorithm for Inverse Problems
Recently it has been shown that using diffusion models for inverse problems can lead to remarkable results. However, these approaches require a closed-form expression of the degradation model and can not support complex degradations. To overcome this limitation, we propose a method (INDigo) that combines invertible neural networks (INN) and diffusion models for general inverse problems. Specifically, we train the forward process of INN to simulate an arbitrary degradation process and use the inverse as a reconstruction process. During the diffusion sampling process, we impose an additional data-consistency step that minimizes the distance between the intermediate result and the INN-optimized result at every iteration, where the INN-optimized image is composed of the coarse information given by the observed degraded image and the details generated by the diffusion process. With the help of INN, our algorithm effectively estimates the details lost in the degradation process and is no longer limited by the requirement of knowing the closed-form expression of the degradation model. Experiments demonstrate that our algorithm obtains competitive results compared with recently leading methods both quantitatively and visually. Moreover, our algorithm performs well on more complex degradation models and real-world low-quality images.
☆ Continual Learning with Pretrained Backbones by Tuning in the Input Space
The intrinsic difficulty in adapting deep learning models to non-stationary environments limits the applicability of neural networks to real-world tasks. This issue is critical in practical supervised learning settings, such as the ones in which a pre-trained model computes projections toward a latent space where different task predictors are sequentially learned over time. As a matter of fact, incrementally fine-tuning the whole model to better adapt to new tasks usually results in catastrophic forgetting, with decreasing performance over the past experiences and losing valuable knowledge from the pre-training stage. In this paper, we propose a novel strategy to make the fine-tuning procedure more effective, by avoiding to update the pre-trained part of the network and learning not only the usual classification head, but also a set of newly-introduced learnable parameters that are responsible for transforming the input data. This process allows the network to effectively leverage the pre-training knowledge and find a good trade-off between plasticity and stability with modest computational efforts, thus especially suitable for on-the-edge settings. Our experiments on four image classification problems in a continual learning setting confirm the quality of the proposed approach when compared to several fine-tuning procedures and to popular continual learning methods.
☆ Human Spine Motion Capture using Perforated Kinesiology Tape
In this work, we present a marker-based multi-view spine tracking method that is specifically adjusted to the requirements for movements in sports. A maximal focus is on the accurate detection of markers and fast usage of the system. For this task, we take advantage of the prior knowledge of the arrangement of dots in perforated kinesiology tape. We detect the tape and its dots using a Mask R-CNN and a blob detector. Here, we can focus on detection only while skipping any image-based feature encoding or matching. We conduct a reasoning in 3D by a linear program and Markov random fields, in which the structure of the kinesiology tape is modeled and the shape of the spine is optimized. In comparison to state-of-the-art systems, we demonstrate that our system achieves high precision and marker density, is robust against occlusions, and capable of capturing fast movements.
☆ Weakly-Supervised Conditional Embedding for Referred Visual Search
This paper presents a new approach to image similarity search in the context of fashion, a domain with inherent ambiguity due to the multiple ways in which images can be considered similar. We introduce the concept of Referred Visual Search (RVS), where users provide additional information to define the desired similarity. We present a new dataset, LAION-RVS-Fashion, consisting of 272K fashion products with 842K images extracted from LAION, designed explicitly for this task. We then propose an innovative method for learning conditional embeddings using weakly-supervised training, achieving a 6% increase in Recall at one (R@1) against a gallery with 2M distractors, compared to classical approaches based on explicit attention and filtering. The proposed method demonstrates robustness, maintaining similar R@1 when dealing with 2.5 times as many distractors as the baseline methods. We believe this is a step forward in the emerging field of Referred Visual Search both in terms of accessible data and approach. Code, data and models are available at https://www.github.com/Simon-Lepage/CondViT-LRVSF .
comment: 20 pages, 13 figures, 4 tables
☆ Zero shot framework for satellite image restoration
Satellite images are typically subject to multiple distortions. Different factors affect the quality of satellite images, including changes in atmosphere, surface reflectance, sun illumination, viewing geometries etc., limiting its application to downstream tasks. In supervised networks, the availability of paired datasets is a strong assumption. Consequently, many unsupervised algorithms have been proposed to address this problem. These methods synthetically generate a large dataset of degraded images using image formation models. A neural network is then trained with an adversarial loss to discriminate between images from distorted and clean domains. However, these methods yield suboptimal performance when tested on real images that do not necessarily conform to the generation mechanism. Also, they require a large amount of training data and are rendered unsuitable when only a few images are available. We propose a distortion disentanglement and knowledge distillation framework for satellite image restoration to address these important issues. Our algorithm requires only two images: the distorted satellite image to be restored and a reference image with similar semantics. Specifically, we first propose a mechanism to disentangle distortion. This enables us to generate images with varying degrees of distortion using the disentangled distortion and the reference image. We then propose the use of knowledge distillation to train a restoration network using the generated image pairs. As a final step, the distorted image is passed through the restoration network to get the final output. Ablation studies show that our proposed mechanism successfully disentangles distortion.
☆ Unsupervised haze removal from underwater images
Several supervised networks exist that remove haze information from underwater images using paired datasets and pixel-wise loss functions. However, training these networks requires large amounts of paired data which is cumbersome, complex and time-consuming. Also, directly using adversarial and cycle consistency loss functions for unsupervised learning is inaccurate as the underlying mapping from clean to underwater images is one-to-many, resulting in an inaccurate constraint on the cycle consistency loss. To address these issues, we propose a new method to remove haze from underwater images using unpaired data. Our model disentangles haze and content information from underwater images using a Haze Disentanglement Network (HDN). The disentangled content is used by a restoration network to generate a clean image using adversarial losses. The disentangled haze is then used as a guide for underwater image regeneration resulting in a strong constraint on cycle consistency loss and improved performance gains. Different ablation studies show that the haze and content from underwater images are effectively separated. Exhaustive experiments reveal that accurate cycle consistency constraint and the proposed network architecture play an important role in yielding enhanced results. Experiments on UFO-120, UWNet, UWScenes, and UIEB underwater datasets indicate that the results of our method outperform prior art both visually and quantitatively.
☆ Instruct-Video2Avatar: Video-to-Avatar Generation with Instructions
We propose a method for synthesizing edited photo-realistic digital avatars with text instructions. Given a short monocular RGB video and text instructions, our method uses an image-conditioned diffusion model to edit one head image and uses the video stylization method to accomplish the editing of other head images. Through iterative training and update (three times or more), our method synthesizes edited photo-realistic animatable 3D neural head avatars with a deformable neural radiance field head synthesis method. In quantitative and qualitative studies on various subjects, our method outperforms state-of-the-art methods.
comment: https://github.com/lsx0101/Instruct-Video2Avatar
☆ A Vessel-Segmentation-Based CycleGAN for Unpaired Multi-modal Retinal Image Synthesis
Unpaired image-to-image translation of retinal images can efficiently increase the training dataset for deep-learning-based multi-modal retinal registration methods. Our method integrates a vessel segmentation network into the image-to-image translation task by extending the CycleGAN framework. The segmentation network is inserted prior to a UNet vision transformer generator network and serves as a shared representation between both domains. We reformulate the original identity loss to learn the direct mapping between the vessel segmentation and the real image. Additionally, we add a segmentation loss term to ensure shared vessel locations between fake and real images. In the experiments, our method shows a visually realistic look and preserves the vessel structures, which is a prerequisite for generating multi-modal training data for image registration.
comment: Accepted to BVM 2023
☆ Robust Fiber ODF Estimation Using Deep Constrained Spherical Deconvolution for Diffusion MRI
Diffusion-weighted magnetic resonance imaging (DW-MRI) is a critical imaging method for capturing and modeling tissue microarchitecture at a millimeter scale. A common practice to model the measured DW-MRI signal is via fiber orientation distribution function (fODF). This function is the essential first step for the downstream tractography and connectivity analyses. With recent advantages in data sharing, large-scale multi-site DW-MRI datasets are being made available for multi-site studies. However, measurement variabilities (e.g., inter- and intra-site variability, hardware performance, and sequence design) are inevitable during the acquisition of DW-MRI. Most existing model-based methods (e.g., constrained spherical deconvolution (CSD)) and learning based methods (e.g., deep learning (DL)) do not explicitly consider such variabilities in fODF modeling, which consequently leads to inferior performance on multi-site and/or longitudinal diffusion studies. In this paper, we propose a novel data-driven deep constrained spherical deconvolution method to explicitly constrain the scan-rescan variabilities for a more reproducible and robust estimation of brain microstructure from repeated DW-MRI scans. Specifically, the proposed method introduces a new 3D volumetric scanner-invariant regularization scheme during the fODF estimation. We study the Human Connectome Project (HCP) young adults test-retest group as well as the MASiVar dataset (with inter- and intra-site scan/rescan data). The Baltimore Longitudinal Study of Aging (BLSA) dataset is employed for external validation. From the experimental results, the proposed data-driven framework outperforms the existing benchmarks in repeated fODF estimation. The proposed method is assessing the downstream connectivity analysis and shows increased performance in distinguishing subjects with different biomarkers.
comment: 33 pages, 7 figures
☆ Towards Unified Text-based Person Retrieval: A Large-scale Multi-Attribute and Language Search Benchmark
In this paper, we introduce a large Multi-Attribute and Language Search dataset for text-based person retrieval, called MALS, and explore the feasibility of performing pre-training on both attribute recognition and image-text matching tasks in one stone. In particular, MALS contains 1,510,330 image-text pairs, which is about 37.5 times larger than prevailing CUHK-PEDES, and all images are annotated with 27 attributes. Considering the privacy concerns and annotation costs, we leverage the off-the-shelf diffusion models to generate the dataset. To verify the feasibility of learning from the generated data, we develop a new joint Attribute Prompt Learning and Text Matching Learning (APTM) framework, considering the shared knowledge between attribute and text. As the name implies, APTM contains an attribute prompt learning stream and a text matching learning stream. (1) The attribute prompt learning leverages the attribute prompts for image-attribute alignment, which enhances the text matching learning. (2) The text matching learning facilitates the representation learning on fine-grained details, and in turn, boosts the attribute prompt learning. Extensive experiments validate the effectiveness of the pre-training on MALS, achieving state-of-the-art retrieval performance via APTM on three challenging real-world benchmarks. In particular, APTM achieves a consistent improvement of +6.60%, +7.39%, and +15.90% Recall@1 accuracy on CUHK-PEDES, ICFG-PEDES, and RSTPReid datasets by a clear margin, respectively.
☆ Image Reconstruction for Accelerated MR Scan with Faster Fourier Convolutional Neural Networks
Partial scan is a common approach to accelerate Magnetic Resonance Imaging (MRI) data acquisition in both 2D and 3D settings. However, accurately reconstructing images from partial scan data (i.e., incomplete k-space matrices) remains challenging due to lack of an effectively global receptive field in both spatial and k-space domains. To address this problem, we propose the following: (1) a novel convolutional operator called Faster Fourier Convolution (FasterFC) to replace the two consecutive convolution operations typically used in convolutional neural networks (e.g., U-Net, ResNet). Based on the spectral convolution theorem in Fourier theory, FasterFC employs alternating kernels of size 1 in 3D case) in different domains to extend the dual-domain receptive field to the global and achieves faster calculation speed than traditional Fast Fourier Convolution (FFC). (2) A 2D accelerated MRI method, FasterFC-End-to-End-VarNet, which uses FasterFC to improve the sensitivity maps and reconstruction quality. (3) A multi-stage 3D accelerated MRI method called FasterFC-based Single-to-group Network (FAS-Net) that utilizes a single-to-group algorithm to guide k-space domain reconstruction, followed by FasterFC-based cascaded convolutional neural networks to expand the effective receptive field in the dual-domain. Experimental results on the fastMRI and Stanford MRI Data datasets demonstrate that FasterFC improves the quality of both 2D and 3D reconstruction. Moreover, FAS-Net, as a 3D high-resolution multi-coil (eight) accelerated MRI method, achieves superior reconstruction performance in both qualitative and quantitative results compared with state-of-the-art 2D and 3D methods.
☆ Unsupervised network for low-light enhancement
Supervised networks address the task of low-light enhancement using paired images. However, collecting a wide variety of low-light/clean paired images is tedious as the scene needs to remain static during imaging. In this paper, we propose an unsupervised low-light enhancement network using contextguided illumination-adaptive norm (CIN). Inspired by coarse to fine methods, we propose to address this task in two stages. In stage-I, a pixel amplifier module (PAM) is used to generate a coarse estimate with an overall improvement in visibility and aesthetic quality. Stage-II further enhances the saturated dark pixels and scene properties of the image using CIN. Different ablation studies show the importance of PAM and CIN in improving the visible quality of the image. Next, we propose a region-adaptive single input multiple output (SIMO) model that can generate multiple enhanced images from a single lowlight image. The objective of SIMO is to let users choose the image of their liking from a pool of enhanced images. Human subjective analysis of SIMO results shows that the distribution of preferred images varies, endorsing the importance of SIMO-type models. Lastly, we propose a low-light road scene (LLRS) dataset having an unpaired collection of low-light and clean scenes. Unlike existing datasets, the clean and low-light scenes in LLRS are real and captured using fixed camera settings. Exhaustive comparisons on publicly available datasets, and the proposed dataset reveal that the results of our model outperform prior art quantitatively and qualitatively.
☆ Single-Stage 3D Geometry-Preserving Depth Estimation Model Training on Dataset Mixtures with Uncalibrated Stereo Data
Nowadays, robotics, AR, and 3D modeling applications attract considerable attention to single-view depth estimation (SVDE) as it allows estimating scene geometry from a single RGB image. Recent works have demonstrated that the accuracy of an SVDE method hugely depends on the diversity and volume of the training data. However, RGB-D datasets obtained via depth capturing or 3D reconstruction are typically small, synthetic datasets are not photorealistic enough, and all these datasets lack diversity. The large-scale and diverse data can be sourced from stereo images or stereo videos from the web. Typically being uncalibrated, stereo data provides disparities up to unknown shift (geometrically incomplete data), so stereo-trained SVDE methods cannot recover 3D geometry. It was recently shown that the distorted point clouds obtained with a stereo-trained SVDE method can be corrected with additional point cloud modules (PCM) separately trained on the geometrically complete data. On the contrary, we propose GP$^{2}$, General-Purpose and Geometry-Preserving training scheme, and show that conventional SVDE models can learn correct shifts themselves without any post-processing, benefiting from using stereo data even in the geometry-preserving setting. Through experiments on different dataset mixtures, we prove that GP$^{2}$-trained models outperform methods relying on PCM in both accuracy and speed, and report the state-of-the-art results in the general-purpose geometry-preserving SVDE. Moreover, we show that SVDE models can learn to predict geometrically correct depth even when geometrically complete data comprises the minor part of the training set.
☆ Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding
We present Video-LLaMA, a multi-modal framework that empowers Large Language Models (LLMs) with the capability of understanding both visual and auditory content in the video. Video-LLaMA bootstraps cross-modal training from the frozen pre-trained visual \& audio encoders and the frozen LLMs. Unlike previous vision- LLMs that focus on static image comprehensions such as MiniGPT-4~\citep{zhu2023minigpt} and LLaVA~\citep{liu2023visualit}, Video-LLaMA tackles two challenges in video understanding: (1) capturing the temporal changes in visual scenes, (2) integrating audio-visual signals. For the first challenge, we propose Video Q-former to extend the pre-trained image encoder to a video encoder and introduce a video-to-text generation task to learn video-language correspondence. For the second challenge, we leverage ImageBind~\citep{girdhar2023imagebind} as the pre-trained audio encoder which performs exceptionally well in aligning different modalities to a common embedding space. And then introduce an Audio Q-former to learn auditory query tokens. To align the output of both visual \& audio encoder with LLM's embedding space, we train Video-LLaMA on a large-scale vision caption dataset and a hign-quantity vision-instruction-tuning dataset. We found Video-LLaMA showcases the ability to perceive and comprehend video content, generating meaningful responses that are grounded in the visual and auditory information present in the videos. This highlights the potential of Video-LLaMA as a promising prototype for audio-visual AI assistants. Our code, pre-trained model, and demo are available at \url{https://github.com/DAMO-NLP-SG/Video-LLaMA}.
comment: Technical Report
☆ Asymmetric Patch Sampling for Contrastive Learning
Asymmetric appearance between positive pair effectively reduces the risk of representation degradation in contrastive learning. However, there are still a mass of appearance similarities between positive pair constructed by the existing methods, which inhibits the further representation improvement. In this paper, we propose a novel asymmetric patch sampling strategy for contrastive learning, to further boost the appearance asymmetry for better representations. Specifically, dual patch sampling strategies are applied to the given image, to obtain asymmetric positive pairs. First, sparse patch sampling is conducted to obtain the first view, which reduces spatial redundancy of image and allows a more asymmetric view. Second, a selective patch sampling is proposed to construct another view with large appearance discrepancy relative to the first one. Due to the inappreciable appearance similarity between positive pair, the trained model is encouraged to capture the similarity on semantics, instead of low-level ones. Experimental results demonstrate that our proposed method significantly outperforms the existing self-supervised methods on both ImageNet-1K and CIFAR dataset, e.g., 2.5% finetune accuracy improvement on CIFAR100. Furthermore, our method achieves state-of-the-art performance on downstream tasks, object detection and instance segmentation on COCO.Additionally, compared to other self-supervised methods, our method is more efficient on both memory and computation during training. The source code is available at https://github.com/visresearch/aps.
☆ Scene as Occupancy
Human driver can easily describe the complex traffic scene by visual system. Such an ability of precise perception is essential for driver's planning. To achieve this, a geometry-aware representation that quantizes the physical 3D scene into structured grid map with semantic labels per cell, termed as 3D Occupancy, would be desirable. Compared to the form of bounding box, a key insight behind occupancy is that it could capture the fine-grained details of critical obstacles in the scene, and thereby facilitate subsequent tasks. Prior or concurrent literature mainly concentrate on a single scene completion task, where we might argue that the potential of this occupancy representation might obsess broader impact. In this paper, we propose OccNet, a multi-view vision-centric pipeline with a cascade and temporal voxel decoder to reconstruct 3D occupancy. At the core of OccNet is a general occupancy embedding to represent 3D physical world. Such a descriptor could be applied towards a wide span of driving tasks, including detection, segmentation and planning. To validate the effectiveness of this new representation and our proposed algorithm, we propose OpenOcc, the first dense high-quality 3D occupancy benchmark built on top of nuScenes. Empirical experiments show that there are evident performance gain across multiple tasks, e.g., motion planning could witness a collision rate reduction by 15%-58%, demonstrating the superiority of our method.
☆ TRACE: 5D Temporal Regression of Avatars with Dynamic Cameras in 3D Environments CVPR2023
Although the estimation of 3D human pose and shape (HPS) is rapidly progressing, current methods still cannot reliably estimate moving humans in global coordinates, which is critical for many applications. This is particularly challenging when the camera is also moving, entangling human and camera motion. To address these issues, we adopt a novel 5D representation (space, time, and identity) that enables end-to-end reasoning about people in scenes. Our method, called TRACE, introduces several novel architectural components. Most importantly, it uses two new "maps" to reason about the 3D trajectory of people over time in camera, and world, coordinates. An additional memory unit enables persistent tracking of people even during long occlusions. TRACE is the first one-stage method to jointly recover and track 3D humans in global coordinates from dynamic cameras. By training it end-to-end, and using full image information, TRACE achieves state-of-the-art performance on tracking and HPS benchmarks. The code and dataset are released for research purposes.
comment: TRACE will appear in CVPR2023
☆ HireVAE: An Online and Adaptive Factor Model Based on Hierarchical and Regime-Switch VAE IJCAI 2023
Factor model is a fundamental investment tool in quantitative investment, which can be empowered by deep learning to become more flexible and efficient in practical complicated investing situations. However, it is still an open question to build a factor model that can conduct stock prediction in an online and adaptive setting, where the model can adapt itself to match the current market regime identified based on only point-in-time market information. To tackle this problem, we propose the first deep learning based online and adaptive factor model, HireVAE, at the core of which is a hierarchical latent space that embeds the underlying relationship between the market situation and stock-wise latent factors, so that HireVAE can effectively estimate useful latent factors given only historical market information and subsequently predict accurate stock returns. Across four commonly used real stock market benchmarks, the proposed HireVAE demonstrate superior performance in terms of active returns over previous methods, verifying the potential of such online and adaptive factor model.
comment: Accepted to IJCAI 2023
Transformer-Based UNet with Multi-Headed Cross-Attention Skip Connections to Eliminate Artifacts in Scanned Documents
The extraction of text in high quality is essential for text-based document analysis tasks like Document Classification or Named Entity Recognition. Unfortunately, this is not always ensured, as poor scan quality and the resulting artifacts lead to errors in the Optical Character Recognition (OCR) process. Current approaches using Convolutional Neural Networks show promising results for background removal tasks but fail correcting artifacts like pixelation or compression errors. For general images, Transformer backbones are getting integrated more frequently in well-known neural network structures for denoising tasks. In this work, a modified UNet structure using a Swin Transformer backbone is presented to remove typical artifacts in scanned documents. Multi-headed cross-attention skip connections are used to more selectively learn features in respective levels of abstraction. The performance of this approach is examined regarding compression errors, pixelation and random noise. An improvement in text extraction quality with a reduced error rate of up to 53.9% on the synthetic data is archived. The pretrained base-model can be easily adapted to new artifacts. The cross-attention skip connections allow to integrate textual information extracted from the encoder or in form of commands to more selectively control the models outcome. The latter is shown by means of an example application.
☆ Using Multiple Dermoscopic Photographs of One Lesion Improves Melanoma Classification via Deep Learning: A Prognostic Diagnostic Accuracy Study
Background: Convolutional neural network (CNN)-based melanoma classifiers face several challenges that limit their usefulness in clinical practice. Objective: To investigate the impact of multiple real-world dermoscopic views of a single lesion of interest on a CNN-based melanoma classifier. Methods: This study evaluated 656 suspected melanoma lesions. Classifier performance was measured using area under the receiver operating characteristic curve (AUROC), expected calibration error (ECE) and maximum confidence change (MCC) for (I) a single-view scenario, (II) a multiview scenario using multiple artificially modified images per lesion and (III) a multiview scenario with multiple real-world images per lesion. Results: The multiview approach with real-world images significantly increased the AUROC from 0.905 (95% CI, 0.879-0.929) in the single-view approach to 0.930 (95% CI, 0.909-0.951). ECE and MCC also improved significantly from 0.131 (95% CI, 0.105-0.159) to 0.072 (95% CI: 0.052-0.093) and from 0.149 (95% CI, 0.125-0.171) to 0.115 (95% CI: 0.099-0.131), respectively. Comparing multiview real-world to artificially modified images showed comparable diagnostic accuracy and uncertainty estimation, but significantly worse robustness for the latter. Conclusion: Using multiple real-world images is an inexpensive method to positively impact the performance of a CNN-based melanoma classifier.
☆ Reassembling Broken Objects using Breaking Curves 3DV
Reassembling 3D broken objects is a challenging task. A robust solution that generalizes well must deal with diverse patterns associated with different types of broken objects. We propose a method that tackles the pairwise assembly of 3D point clouds, that is agnostic on the type of object, and that relies solely on their geometrical information, without any prior information on the shape of the reconstructed object. The method receives two point clouds as input and segments them into regions using detected closed boundary contours, known as breaking curves. Possible alignment combinations of the regions of each broken object are evaluated and the best one is selected as the final alignment. Experiments were carried out both on available 3D scanned objects and on a recent benchmark for synthetic broken objects. Results show that our solution performs well in reassembling different kinds of broken objects.
comment: 4 pages, accepted at 3DVR Workshop @ CVPR 2023
☆ Cheap-fake Detection with LLM using Prompt Engineering ICME2023
The misuse of real photographs with conflicting image captions in news items is an example of the out-of-context (OOC) misuse of media. In order to detect OOC media, individuals must determine the accuracy of the statement and evaluate whether the triplet (~\textit{i.e.}, the image and two captions) relates to the same event. This paper presents a novel learnable approach for detecting OOC media in ICME'23 Grand Challenge on Detecting Cheapfakes. The proposed method is based on the COSMOS structure, which assesses the coherence between an image and captions, as well as between two captions. We enhance the baseline algorithm by incorporating a Large Language Model (LLM), GPT3.5, as a feature extractor. Specifically, we propose an innovative approach to feature extraction utilizing prompt engineering to develop a robust and reliable feature extractor with GPT3.5 model. The proposed method captures the correlation between two captions and effectively integrates this module into the COSMOS baseline model, which allows for a deeper understanding of the relationship between captions. By incorporating this module, we demonstrate the potential for significant improvements in cheap-fakes detection performance. The proposed methodology holds promising implications for various applications such as natural language processing, image captioning, and text-to-image synthesis. Docker for submission is available at https://hub.docker.com/repository/docker/mulns/ acmmmcheapfakes.
comment: ICME2023 Workshop
☆ Differentially Private Cross-camera Person Re-identification
Camera-based person re-identification is a heavily privacy-invading task by design, benefiting from rich visual data to match together person representations across different cameras. This high-dimensional data can then easily be used for other, perhaps less desirable, applications. We here investigate the possibility of protecting such image data against uses outside of the intended re-identification task, and introduce a differential privacy mechanism leveraging both pixelisation and colour quantisation for this purpose. We show its ability to distort images in such a way that adverse task performances are significantly reduced, while retaining high re-identification performances.
☆ STAR Loss: Reducing Semantic Ambiguity in Facial Landmark Detection CVPR 2023
Recently, deep learning-based facial landmark detection has achieved significant improvement. However, the semantic ambiguity problem degrades detection performance. Specifically, the semantic ambiguity causes inconsistent annotation and negatively affects the model's convergence, leading to worse accuracy and instability prediction. To solve this problem, we propose a Self-adapTive Ambiguity Reduction (STAR) loss by exploiting the properties of semantic ambiguity. We find that semantic ambiguity results in the anisotropic predicted distribution, which inspires us to use predicted distribution to represent semantic ambiguity. Based on this, we design the STAR loss that measures the anisotropism of the predicted distribution. Compared with the standard regression loss, STAR loss is encouraged to be small when the predicted distribution is anisotropic and thus adaptively mitigates the impact of semantic ambiguity. Moreover, we propose two kinds of eigenvalue restriction methods that could avoid both distribution's abnormal change and the model's premature convergence. Finally, the comprehensive experiments demonstrate that STAR loss outperforms the state-of-the-art methods on three benchmarks, i.e., COFW, 300W, and WFLW, with negligible computation overhead. Code is at https://github.com/ZhenglinZhou/STAR.
comment: 14 pages, 7 figures, accepted by CVPR 2023
☆ A2B: Anchor to Barycentric Coordinate for Robust Correspondence
There is a long-standing problem of repeated patterns in correspondence problems, where mismatches frequently occur because of inherent ambiguity. The unique position information associated with repeated patterns makes coordinate representations a useful supplement to appearance representations for improving feature correspondences. However, the issue of appropriate coordinate representation has remained unresolved. In this study, we demonstrate that geometric-invariant coordinate representations, such as barycentric coordinates, can significantly reduce mismatches between features. The first step is to establish a theoretical foundation for geometrically invariant coordinates. We present a seed matching and filtering network (SMFNet) that combines feature matching and consistency filtering with a coarse-to-fine matching strategy in order to acquire reliable sparse correspondences. We then introduce DEGREE, a novel anchor-to-barycentric (A2B) coordinate encoding approach, which generates multiple affine-invariant correspondence coordinates from paired images. DEGREE can be used as a plug-in with standard descriptors, feature matchers, and consistency filters to improve the matching quality. Extensive experiments in synthesized indoor and outdoor datasets demonstrate that DEGREE alleviates the problem of repeated patterns and helps achieve state-of-the-art performance. Furthermore, DEGREE also reports competitive performance in the third Image Matching Challenge at CVPR 2021. This approach offers a new perspective to alleviate the problem of repeated patterns and emphasizes the importance of choosing coordinate representations for feature correspondences.
comment: Accepted by International Journal of Computer Vision
☆ Towards Better Explanations for Object Detection
Recent advances in Artificial Intelligence (AI) technology have promoted their use in almost every field. The growing complexity of deep neural networks (DNNs) makes it increasingly difficult and important to explain the inner workings and decisions of the network. However, most current techniques for explaining DNNs focus mainly on interpreting classification tasks. This paper proposes a method to explain the decision for any object detection model called D-CLOSE. To closely track the model's behavior, we used multiple levels of segmentation on the image and a process to combine them. We performed tests on the MS-COCO dataset with the YOLOX model, which shows that our method outperforms D-RISE and can give a better quality and less noise explanation.
comment: 9 pages, 10 figures
☆ ZIGNeRF: Zero-shot 3D Scene Representation with Invertible Generative Neural Radiance Fields
Generative Neural Radiance Fields (NeRFs) have demonstrated remarkable proficiency in synthesizing multi-view images by learning the distribution of a set of unposed images. Despite the aptitude of existing generative NeRFs in generating 3D-consistent high-quality random samples within data distribution, the creation of a 3D representation of a singular input image remains a formidable challenge. In this manuscript, we introduce ZIGNeRF, an innovative model that executes zero-shot Generative Adversarial Network (GAN) inversion for the generation of multi-view images from a single out-of-domain image. The model is underpinned by a novel inverter that maps out-of-domain images into the latent code of the generator manifold. Notably, ZIGNeRF is capable of disentangling the object from the background and executing 3D operations such as 360-degree rotation or depth and horizontal translation. The efficacy of our model is validated using multiple real-image datasets: Cats, AFHQ, CelebA, CelebA-HQ, and CompCars.
☆ Overcoming Weak Visual-Textual Alignment for Video Moment Retrieval
Video moment retrieval (VMR) aims to identify the specific moment in an untrimmed video for a given natural language query. However, this task is prone to suffer the weak visual-textual alignment problem from query ambiguity, potentially limiting further performance gains and generalization capability. Due to the complex multimodal interactions in videos, a query may not fully cover the relevant details of the corresponding moment, and the moment may contain misaligned and irrelevant frames. To tackle this problem, we propose a straightforward yet effective model, called Background-aware Moment DEtection TRansformer (BM-DETR). Given a target query and its moment, BM-DETR also takes negative queries corresponding to different moments. Specifically, our model learns to predict the target moment from the joint probability of the given query and the complement of negative queries for each candidate frame. In this way, it leverages the surrounding background to consider relative importance, improving moment sensitivity. Extensive experiments on Charades-STA and QVHighlights demonstrate the effectiveness of our model. Moreover, we show that BM-DETR can perform robustly in three challenging VMR scenarios, such as several out-of-distribution test cases, demonstrating superior generalization ability.
comment: Under Review; Our code is available at https://github.com/minjoong507/BM-DETR
☆ User-friendly Image Editing with Minimal Text Input: Leveraging Captioning and Injection Techniques
Recent text-driven image editing in diffusion models has shown remarkable success. However, the existing methods assume that the user's description sufficiently grounds the contexts in the source image, such as objects, background, style, and their relations. This assumption is unsuitable for real-world applications because users have to manually engineer text prompts to find optimal descriptions for different images. From the users' standpoint, prompt engineering is a labor-intensive process, and users prefer to provide a target word for editing instead of a full sentence. To address this problem, we first demonstrate the importance of a detailed text description of the source image, by dividing prompts into three categories based on the level of semantic details. Then, we propose simple yet effective methods by combining prompt generation frameworks, thereby making the prompt engineering process more user-friendly. Extensive qualitative and quantitative experiments demonstrate the importance of prompts in text-driven image editing and our method is comparable to ground-truth prompts.
☆ NFTVis: Visual Analysis of NFT Performance
A non-fungible token (NFT) is a data unit stored on the blockchain. Nowadays, more and more investors and collectors (NFT traders), who participate in transactions of NFTs, have an urgent need to assess the performance of NFTs. However, there are two challenges for NFT traders when analyzing the performance of NFT. First, the current rarity models have flaws and are sometimes not convincing. In addition, NFT performance is dependent on multiple factors, such as images (high-dimensional data), history transactions (network), and market evolution (time series). It is difficult to take comprehensive consideration and analyze NFT performance efficiently. To address these challenges, we propose NFTVis, a visual analysis system that facilitates assessing individual NFT performance. A new NFT rarity model is proposed to quantify NFTs with images. Four well-coordinated views are designed to represent the various factors affecting the performance of the NFT. Finally, we evaluate the usefulness and effectiveness of our system using two case studies and user studies.
comment: This manuscript is accepted for publication in Proceedings of the 16th IEEE Pacific Visualization Symposium (PacificVis '23)
☆ Cyclic Learning: Bridging Image-level Labels and Nuclei Instance Segmentation
Nuclei instance segmentation on histopathology images is of great clinical value for disease analysis. Generally, fully-supervised algorithms for this task require pixel-wise manual annotations, which is especially time-consuming and laborious for the high nuclei density. To alleviate the annotation burden, we seek to solve the problem through image-level weakly supervised learning, which is underexplored for nuclei instance segmentation. Compared with most existing methods using other weak annotations (scribble, point, etc.) for nuclei instance segmentation, our method is more labor-saving. The obstacle to using image-level annotations in nuclei instance segmentation is the lack of adequate location information, leading to severe nuclei omission or overlaps. In this paper, we propose a novel image-level weakly supervised method, called cyclic learning, to solve this problem. Cyclic learning comprises a front-end classification task and a back-end semi-supervised instance segmentation task to benefit from multi-task learning (MTL). We utilize a deep learning classifier with interpretability as the front-end to convert image-level labels to sets of high-confidence pseudo masks and establish a semi-supervised architecture as the back-end to conduct nuclei instance segmentation under the supervision of these pseudo masks. Most importantly, cyclic learning is designed to circularly share knowledge between the front-end classifier and the back-end semi-supervised part, which allows the whole system to fully extract the underlying information from image-level labels and converge to a better optimum. Experiments on three datasets demonstrate the good generality of our method, which outperforms other image-level weakly supervised methods for nuclei instance segmentation, and achieves comparable performance to fully-supervised methods.
comment: This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI https://doi.org/10.1109/TMI.2023.3275609, IEEE Transactions on Medical Imaging. Code: https://github.com/wuyongjianCODE/Cyclic
☆ Cross-Modal Vertical Federated Learning for MRI Reconstruction
Federated learning enables multiple hospitals to cooperatively learn a shared model without privacy disclosure. Existing methods often take a common assumption that the data from different hospitals have the same modalities. However, such a setting is difficult to fully satisfy in practical applications, since the imaging guidelines may be different between hospitals, which makes the number of individuals with the same set of modalities limited. To this end, we formulate this practical-yet-challenging cross-modal vertical federated learning task, in which shape data from multiple hospitals have different modalities with a small amount of multi-modality data collected from the same individuals. To tackle such a situation, we develop a novel framework, namely Federated Consistent Regularization constrained Feature Disentanglement (Fed-CRFD), for boosting MRI reconstruction by effectively exploring the overlapping samples (individuals with multi-modalities) and solving the domain shift problem caused by different modalities. Particularly, our Fed-CRFD involves an intra-client feature disentangle scheme to decouple data into modality-invariant and modality-specific features, where the modality-invariant features are leveraged to mitigate the domain shift problem. In addition, a cross-client latent representation consistency constraint is proposed specifically for the overlapping samples to further align the modality-invariant features extracted from different modalities. Hence, our method can fully exploit the multi-source data from hospitals while alleviating the domain shift problem. Extensive experiments on two typical MRI datasets demonstrate that our network clearly outperforms state-of-the-art MRI reconstruction methods. The source code will be publicly released upon the publication of this work.
comment: 12 pages, 7 figures
☆ Calib-Anything: Zero-training LiDAR-Camera Extrinsic Calibration Method Using Segment Anything
The research on extrinsic calibration between Light Detection and Ranging(LiDAR) and camera are being promoted to a more accurate, automatic and generic manner. Since deep learning has been employed in calibration, the restrictions on the scene are greatly reduced. However, data driven method has the drawback of low transfer-ability. It cannot adapt to dataset variations unless additional training is taken. With the advent of foundation model, this problem can be significantly mitigated. By using the Segment Anything Model(SAM), we propose a novel LiDAR-camera calibration method, which requires zero extra training and adapts to common scenes. With an initial guess, we opimize the extrinsic parameter by maximizing the consistency of points that are projected inside each image mask. The consistency includes three properties of the point cloud: the intensity, normal vector and categories derived from some segmentation methods. The experiments on different dataset have demonstrated the generality and comparable accuracy of our method. The code is available at https://github.com/OpenCalib/CalibAnything.
comment: 5 pages, 4 figures
☆ Dynamic Interactive Relation Capturing via Scene Graph Learning for Robotic Surgical Report Generation
For robot-assisted surgery, an accurate surgical report reflects clinical operations during surgery and helps document entry tasks, post-operative analysis and follow-up treatment. It is a challenging task due to many complex and diverse interactions between instruments and tissues in the surgical scene. Although existing surgical report generation methods based on deep learning have achieved large success, they often ignore the interactive relation between tissues and instrumental tools, thereby degrading the report generation performance. This paper presents a neural network to boost surgical report generation by explicitly exploring the interactive relation between tissues and surgical instruments. We validate the effectiveness of our method on a widely-used robotic surgery benchmark dataset, and experimental results show that our network can significantly outperform existing state-of-the-art surgical report generation methods (e.g., 7.48% and 5.43% higher for BLEU-1 and ROUGE).
☆ Continuous Cartesian Genetic Programming based representation for Multi-Objective Neural Architecture Search
We propose a novel approach for the challenge of designing less complex yet highly effective convolutional neural networks (CNNs) through the use of cartesian genetic programming (CGP) for neural architecture search (NAS). Our approach combines real-based and block-chained CNNs representations based on CGP for optimization in the continuous domain using multi-objective evolutionary algorithms (MOEAs). Two variants are introduced that differ in the granularity of the search space they consider. The proposed CGP-NASV1 and CGP-NASV2 algorithms were evaluated using the non-dominated sorting genetic algorithm II (NSGA-II) on the CIFAR-10 and CIFAR-100 datasets. The empirical analysis was extended to assess the crossover operator from differential evolution (DE), the multi-objective evolutionary algorithm based on decomposition (MOEA/D) and S metric selection evolutionary multi-objective algorithm (SMS-EMOA) using the same representation. Experimental results demonstrate that our approach is competitive with state-of-the-art proposals in terms of classification performance and model complexity.
☆ Learned Alternating Minimization Algorithm for Dual-domain Sparse-View CT Reconstruction
We propose a novel Learned Alternating Minimization Algorithm (LAMA) for dual-domain sparse-view CT image reconstruction. LAMA is naturally induced by a variational model for CT reconstruction with learnable nonsmooth nonconvex regularizers, which are parameterized as composite functions of deep networks in both image and sinogram domains. To minimize the objective of the model, we incorporate the smoothing technique and residual learning architecture into the design of LAMA. We show that LAMA substantially reduces network complexity, improves memory efficiency and reconstruction accuracy, and is provably convergent for reliable reconstructions. Extensive numerical experiments demonstrate that LAMA outperforms existing methods by a wide margin on multiple benchmark CT datasets.
☆ Computational 3D topographic microscopy from terabytes of data per sample
We present a large-scale computational 3D topographic microscope that enables 6-gigapixel profilometric 3D imaging at micron-scale resolution across $>$110 cm$^2$ areas over multi-millimeter axial ranges. Our computational microscope, termed STARCAM (Scanning Topographic All-in-focus Reconstruction with a Computational Array Microscope), features a parallelized, 54-camera architecture with 3-axis translation to capture, for each sample of interest, a multi-dimensional, 2.1-terabyte (TB) dataset, consisting of a total of 224,640 9.4-megapixel images. We developed a self-supervised neural network-based algorithm for 3D reconstruction and stitching that jointly estimates an all-in-focus photometric composite and 3D height map across the entire field of view, using multi-view stereo information and image sharpness as a focal metric. The memory-efficient, compressed differentiable representation offered by the neural network effectively enables joint participation of the entire multi-TB dataset during the reconstruction process. To demonstrate the broad utility of our new computational microscope, we applied STARCAM to a variety of decimeter-scale objects, with applications ranging from cultural heritage to industrial inspection.
☆ Do-GOOD: Towards Distribution Shift Evaluation for Pre-Trained Visual Document Understanding Models SIGIR 2023
Numerous pre-training techniques for visual document understanding (VDU) have recently shown substantial improvements in performance across a wide range of document tasks. However, these pre-trained VDU models cannot guarantee continued success when the distribution of test data differs from the distribution of training data. In this paper, to investigate how robust existing pre-trained VDU models are to various distribution shifts, we first develop an out-of-distribution (OOD) benchmark termed Do-GOOD for the fine-Grained analysis on Document image-related tasks specifically. The Do-GOOD benchmark defines the underlying mechanisms that result in different distribution shifts and contains 9 OOD datasets covering 3 VDU related tasks, e.g., document information extraction, classification and question answering. We then evaluate the robustness and perform a fine-grained analysis of 5 latest VDU pre-trained models and 2 typical OOD generalization algorithms on these OOD datasets. Results from the experiments demonstrate that there is a significant performance gap between the in-distribution (ID) and OOD settings for document images, and that fine-grained analysis of distribution shifts can reveal the brittle nature of existing pre-trained VDU models and OOD generalization algorithms. The code and datasets for our Do-GOOD benchmark can be found at https://github.com/MAEHCM/Do-GOOD.
comment: SIGIR 2023. The code and datasets for our Do-GOOD benchmark can be found at https://github.com/MAEHCM/Do-GOOD
☆ ReContrast: Domain-Specific Anomaly Detection via Contrastive Reconstruction
Most advanced unsupervised anomaly detection (UAD) methods rely on modeling feature representations of frozen encoder networks pre-trained on large-scale datasets, e.g. ImageNet. However, the features extracted from the encoders that are borrowed from natural image domains coincide little with the features required in the target UAD domain, such as industrial inspection and medical imaging. In this paper, we propose a novel epistemic UAD method, namely ReContrast, which optimizes the entire network to reduce biases towards the pre-trained image domain and orients the network in the target domain. We start with a feature reconstruction approach that detects anomalies from errors. Essentially, the elements of contrastive learning are elegantly embedded in feature reconstruction to prevent the network from training instability, pattern collapse, and identical shortcut, while simultaneously optimizing both the encoder and decoder on the target domain. To demonstrate our transfer ability on various image domains, we conduct extensive experiments across two popular industrial defect detection benchmarks and three medical image UAD tasks, which shows our superiority over current state-of-the-art methods.
comment: under review
☆ A Novel Interpretable and Generalizable Re-synchronization Model for Cued Speech based on a Multi-Cuer Corpus INTERSPEECH2023
Cued Speech (CS) is a multi-modal visual coding system combining lip reading with several hand cues at the phonetic level to make the spoken language visible to the hearing impaired. Previous studies solved asynchronous problems between lip and hand movements by a cuer\footnote{The people who perform Cued Speech are called the cuer.}-dependent piecewise linear model for English and French CS. In this work, we innovatively propose three statistical measure on the lip stream to build an interpretable and generalizable model for predicting hand preceding time (HPT), which achieves cuer-independent by a proper normalization. Particularly, we build the first Mandarin CS corpus comprising annotated videos from five speakers including three normal and two hearing impaired individuals. Consequently, we show that the hand preceding phenomenon exists in Mandarin CS production with significant differences between normal and hearing impaired people. Extensive experiments demonstrate that our model outperforms the baseline and the previous state-of-the-art methods.
comment: 5 pages, 4 figures, Accepted to INTERSPEECH2023
☆ DAGrid: Directed Accumulator Grid
Recent research highlights that the Directed Accumulator (DA), through its parametrization of geometric priors into neural networks, has notably improved the performance of medical image recognition, particularly with small and imbalanced datasets. However, DA's potential in pixel-wise dense predictions is unexplored. To bridge this gap, we present the Directed Accumulator Grid (DAGrid), which allows geometric-preserving filtering in neural networks, thus broadening the scope of DA's applications to include pixel-level dense prediction tasks. DAGrid utilizes homogeneous data types in conjunction with designed sampling grids to construct geometrically transformed representations, retaining intricate geometric information and promoting long-range information propagation within the neural networks. Contrary to its symmetric counterpart, grid sampling, which might lose information in the sampling process, DAGrid aggregates all pixels, ensuring a comprehensive representation in the transformed space. The parallelization of DAGrid on modern GPUs is facilitated using CUDA programming, and also back propagation is enabled for deep neural network training. Empirical results show DAGrid-enhanced neural networks excel in supervised skin lesion segmentation and unsupervised cardiac image registration. Specifically, the network incorporating DAGrid has realized a 70.8% reduction in network parameter size and a 96.8% decrease in FLOPs, while concurrently improving the Dice score for skin lesion segmentation by 1.0% compared to state-of-the-art transformers. Furthermore, it has achieved improvements of 4.4% and 8.2% in the average Dice score and Dice score of the left ventricular mass, respectively, indicating an increase in registration accuracy for cardiac images. The source code is available at https://github.com/tinymilky/DeDA.
comment: 19 pages, 9 figures, 4 tables
☆ MotionTrack: Learning Motion Predictor for Multiple Object Tracking
Significant advancements have been made in multi-object tracking (MOT) with the development of detection and re-identification (ReID) techniques. Despite these developments, the task of accurately tracking objects in scenarios with homogeneous appearance and heterogeneous motion remains challenging due to the insufficient discriminability of ReID features and the predominant use of linear motion models in MOT. In this context, we present a novel learnable motion predictor, named MotionTrack, which comprehensively incorporates two levels of granularity of motion features to enhance the modeling of temporal dynamics and facilitate accurate future motion prediction of individual objects. Specifically, the proposed approach adopts a self-attention mechanism to capture token-level information and a Dynamic MLP layer to model channel-level features. MotionTrack is a simple, online tracking approach. Our experimental results demonstrate that MotionTrack yields state-of-the-art performance on demanding datasets such as SportsMOT and Dancetrack, which feature highly nonlinear object motion. Notably, without fine-tuning on target datasets, MotionTrack also exhibits competitive performance on conventional benchmarks including MOT17 and MOT20.
☆ Stable Diffusion is Untable
Recently, text-to-image models have been thriving. Despite their powerful generative capacity, our research has uncovered a lack of robustness in this generation process. Specifically, the introduction of small perturbations to the text prompts can result in the blending of primary subjects with other categories or their complete disappearance in the generated images. In this paper, we propose Auto-attack on Text-to-image Models (ATM), a gradient-based approach, to effectively and efficiently generate such perturbations. By learning a Gumbel Softmax distribution, we can make the discrete process of word replacement or extension continuous, thus ensuring the differentiability of the perturbation generation. Once the distribution is learned, ATM can sample multiple attack samples simultaneously. These attack samples can prevent the generative model from generating the desired subjects without compromising image quality. ATM has achieved a 91.1% success rate in short-text attacks and an 81.2% success rate in long-text attacks. Further empirical analysis revealed four attack patterns based on: 1) the variability in generation speed, 2) the similarity of coarse-grained characteristics, 3) the polysemy of words, and 4) the positioning of words.
comment: 22 pages, 20 figures
☆ Learning from Noisy Labels Generated by Extremely Point Annotations for OCT Fluid Segmentation
Automatic segmentation of fluid in OCT (Optical Coherence Tomography) images is beneficial for ophthalmologists to make an accurate diagnosis. Currently, data-driven convolutional neural networks (CNNs) have achieved great success in OCT fluid segmentation. However, obtaining pixel-level masks of OCT images is time-consuming and requires expertise. The popular weakly-supervised strategy is to generate noisy pseudo-labels from weak annotations, but the noise information introduced may mislead the model training. To address this issue, (i) we propose a superpixel-guided method for generating noisy labels from weak point annotations, called Point to Noisy by Superpixel (PNS), which limits the network from over-fitting noise by assigning low confidence to suspiciously noisy label pixels, and (ii) we propose a Two-Stage Mean-Teacher-assisted Confident Learning (2SMTCL) method based on MTCL for multi-category OCT fluid segmentation, which alleviates the uncertainty and computing power consumption introduced by the real-time characterization noise of MTCL. For evaluation, we have constructed a 2D OCT fluid segmentation dataset. Compared with other state-of-art label-denoising methods, comprehensive experimental results demonstrate that the proposed method can achieve excellent performance in OCT fluid segmentation as well as label denoising. Our study provides an efficient, accurate, and practical solution for fluid segmentation of OCT images, which is expected to have a positive impact on the diagnosis and treatment of patients in the field of ophthalmology.
comment: Submission to IEEE Transactions on Biomedical Engineering
☆ Exploring the Role of the Bottleneck in Slot-Based Models Through Covariance Regularization
In this project we attempt to make slot-based models with an image reconstruction objective competitive with those that use a feature reconstruction objective on real world datasets. We propose a loss-based approach to constricting the bottleneck of slot-based models, allowing larger-capacity encoder networks to be used with Slot Attention without producing degenerate stripe-shaped masks. We find that our proposed method offers an improvement over the baseline Slot Attention model but does not reach the performance of \dinosaur on the COCO2017 dataset. Throughout this project, we confirm the superiority of a feature reconstruction objective over an image reconstruction objective and explore the role of the architectural bottleneck in slot-based models.
comment: 14 pages, 10 figures
☆ Spatial Implicit Neural Representations for Global-Scale Species Mapping ICML 2023
Estimating the geographical range of a species from sparse observations is a challenging and important geospatial prediction problem. Given a set of locations where a species has been observed, the goal is to build a model to predict whether the species is present or absent at any location. This problem has a long history in ecology, but traditional methods struggle to take advantage of emerging large-scale crowdsourced datasets which can include tens of millions of records for hundreds of thousands of species. In this work, we use Spatial Implicit Neural Representations (SINRs) to jointly estimate the geographical range of 47k species simultaneously. We find that our approach scales gracefully, making increasingly better predictions as we increase the number of species and the amount of data per species when training. To make this problem accessible to machine learning researchers, we provide four new benchmarks that measure different aspects of species range estimation and spatial representation learning. Using these benchmarks, we demonstrate that noisy and biased crowdsourced data can be combined with implicit neural representations to approximate expert-developed range maps for many species.
comment: ICML 2023
☆ Video Diffusion Models with Local-Global Context Guidance IJCAI 2023
Diffusion models have emerged as a powerful paradigm in video synthesis tasks including prediction, generation, and interpolation. Due to the limitation of the computational budget, existing methods usually implement conditional diffusion models with an autoregressive inference pipeline, in which the future fragment is predicted based on the distribution of adjacent past frames. However, only the conditions from a few previous frames can't capture the global temporal coherence, leading to inconsistent or even outrageous results in long-term video prediction. In this paper, we propose a Local-Global Context guided Video Diffusion model (LGC-VD) to capture multi-perception conditions for producing high-quality videos in both conditional/unconditional settings. In LGC-VD, the UNet is implemented with stacked residual blocks with self-attention units, avoiding the undesirable computational cost in 3D Conv. We construct a local-global context guidance strategy to capture the multi-perceptual embedding of the past fragment to boost the consistency of future prediction. Furthermore, we propose a two-stage training strategy to alleviate the effect of noisy frames for more stable predictions. Our experiments demonstrate that the proposed method achieves favorable performance on video prediction, interpolation, and unconditional video generation. We release code at https://github.com/exisas/LGC-VD.
comment: Accepted for publication at IJCAI 2023. To appear
☆ Learning from Multi-View Representation for Point-Cloud Pre-Training
A critical problem in the pre-training of 3D point clouds is leveraging massive 2D data. A fundamental challenge is to address the 2D-3D domain gap. This paper proposes a novel approach to point-cloud pre-training that enables learning 3D representations by leveraging pre-trained 2D-based networks. In particular, it avoids overfitting to 2D representations and potentially discarding critical 3D features for 3D recognition tasks. The key to our approach is a novel multi-view representation, which learns a shared 3D feature volume consistent with deep features extracted from multiple 2D camera views. The 2D deep features are regularized using pre-trained 2D networks through the 2D knowledge transfer loss. To prevent the resulting 3D feature representations from discarding 3D signals, we introduce the multi-view consistency loss that forces the projected 2D feature representations to capture pixel-wise correspondences across different views. Such correspondences induce 3D geometry and effectively retain 3D features in the projected 2D features. Experimental results demonstrate that our pre-trained model can be successfully transferred to various downstream tasks, including 3D detection and semantic segmentation, and achieve state-of-the-art performance.
comment: 14 pages, 5 figures
☆ Inflated 3D Convolution-Transformer for Weakly-supervised Carotid Stenosis Grading with Ultrasound Videos MICCAI 2023
Localization of the narrowest position of the vessel and corresponding vessel and remnant vessel delineation in carotid ultrasound (US) are essential for carotid stenosis grading (CSG) in clinical practice. However, the pipeline is time-consuming and tough due to the ambiguous boundaries of plaque and temporal variation. To automatize this procedure, a large number of manual delineations are usually required, which is not only laborious but also not reliable given the annotation difficulty. In this study, we present the first video classification framework for automatic CSG. Our contribution is three-fold. First, to avoid the requirement of laborious and unreliable annotation, we propose a novel and effective video classification network for weakly-supervised CSG. Second, to ease the model training, we adopt an inflation strategy for the network, where pre-trained 2D convolution weights can be adapted into the 3D counterpart in our network. In this way, the existing pre-trained large model can be used as an effective warm start for our network. Third, to enhance the feature discrimination of the video, we propose a novel attention-guided multi-dimension fusion (AMDF) transformer encoder to model and integrate global dependencies within and across spatial and temporal dimensions, where two lightweight cross-dimensional attention mechanisms are designed. Our approach is extensively validated on a large clinically collected carotid US video dataset, demonstrating state-of-the-art performance compared with strong competitors.
comment: Accepted by MICCAI 2023
☆ Fourier Test-time Adaptation with Multi-level Consistency for Robust Classification MICCAI 2023
Deep classifiers may encounter significant performance degradation when processing unseen testing data from varying centers, vendors, and protocols. Ensuring the robustness of deep models against these domain shifts is crucial for their widespread clinical application. In this study, we propose a novel approach called Fourier Test-time Adaptation (FTTA), which employs a dual-adaptation design to integrate input and model tuning, thereby jointly improving the model robustness. The main idea of FTTA is to build a reliable multi-level consistency measurement of paired inputs for achieving self-correction of prediction. Our contribution is two-fold. First, we encourage consistency in global features and local attention maps between the two transformed images of the same input. Here, the transformation refers to Fourier-based input adaptation, which can transfer one unseen image into source style to reduce the domain gap. Furthermore, we leverage style-interpolated images to enhance the global and local features with learnable parameters, which can smooth the consistency measurement and accelerate convergence. Second, we introduce a regularization technique that utilizes style interpolation consistency in the frequency space to encourage self-consistency in the logit space of the model output. This regularization provides strong self-supervised signals for robustness enhancement. FTTA was extensively validated on three large classification datasets with different modalities and organs. Experimental results show that FTTA is general and outperforms other strong state-of-the-art methods.
comment: Accepted by MICCAI 2023
☆ Disaster Anomaly Detector via Deeper FCDDs for Explainable Initial Responses
Urban and rural areas can often be devastated by extreme natural disasters. Towards any disaster event, an initial response is the key to rescuing within 72 hours and prompt recovery. For the stage of initial responses, it is important to quickly recognize the disaster damage over a wide area and determine priority areas. Among machine learning algorithms, deep anomaly detection is effective in detecting devastated features that are different from ordinary vision everyday life. In addition, explainable computer vision applications have been expected to justify the initial responses. In this paper, we propose an anomaly detection application utilizing the deeper fully-convolutional data descriptions (FCDDs), that enables to localize devastated features and visualize damage-marked heatmaps. More concretely, we show numerous training and test results to a dataset AIDER with the four disaster categories: collapsed buildings, traffic accidents, fires, and flooding areas. We also implement ablation studies of anomalous class imbalance and the data scale competing against the normal class. Finally, we discuss future works to improve more robust, explainable applications for effective initial responses.
comment: 10 pages, 12 figures, 9 tables. arXiv admin note: substantial text overlap with arXiv:2305.05103
☆ ICDAR 2023 Competition on Structured Text Extraction from Visually-Rich Document Images ICDAR 2023
Structured text extraction is one of the most valuable and challenging application directions in the field of Document AI. However, the scenarios of past benchmarks are limited, and the corresponding evaluation protocols usually focus on the submodules of the structured text extraction scheme. In order to eliminate these problems, we organized the ICDAR 2023 competition on Structured text extraction from Visually-Rich Document images (SVRD). We set up two tracks for SVRD including Track 1: HUST-CELL and Track 2: Baidu-FEST, where HUST-CELL aims to evaluate the end-to-end performance of Complex Entity Linking and Labeling, and Baidu-FEST focuses on evaluating the performance and generalization of Zero-shot / Few-shot Structured Text extraction from an end-to-end perspective. Compared to the current document benchmarks, our two tracks of competition benchmark enriches the scenarios greatly and contains more than 50 types of visually-rich document images (mainly from the actual enterprise applications). The competition opened on 30th December, 2022 and closed on 24th March, 2023. There are 35 participants and 91 valid submissions received for Track 1, and 15 participants and 26 valid submissions received for Track 2. In this report we will presents the motivation, competition datasets, task definition, evaluation protocol, and submission summaries. According to the performance of the submissions, we believe there is still a large gap on the expected information extraction performance for complex and zero-shot scenarios. It is hoped that this competition will attract many researchers in the field of CV and NLP, and bring some new thoughts to the field of Document AI.
comment: ICDAR 2023 Competition on SVRD report (To be appear in ICDAR 2023)
☆ Dual self-distillation of U-shaped networks for 3D medical image segmentation
U-shaped networks and its variants have demonstrated exceptional results for medical image segmentation. In this paper, we propose a novel dual self-distillation (DSD) framework for U-shaped networks for 3D medical image segmentation. DSD distills knowledge from the ground-truth segmentation labels to the decoder layers and also between the encoder and decoder layers of a single U-shaped network. DSD is a generalized training strategy that could be attached to the backbone architecture of any U-shaped network to further improve its segmentation performance. We attached DSD on two state-of-the-art U-shaped backbones, and extensive experiments on two public 3D medical image segmentation datasets (cardiac substructure and brain tumor) demonstrated significant improvement over those backbones. On average, after attaching DSD to the U-shaped backbones, we observed an improvement of 4.25% and 3.15% in Dice similarity score for cardiac substructure and brain tumor segmentation respectively.
comment: 12 pages, 5 figures, 3 tables
☆ Brain Tumor Recurrence vs. Radiation Necrosis Classification and Patient Survivability Prediction
GBM (Glioblastoma multiforme) is the most aggressive type of brain tumor in adults that has a short survival rate even after aggressive treatment with surgery and radiation therapy. The changes on magnetic resonance imaging (MRI) for patients with GBM after radiotherapy are indicative of either radiation-induced necrosis (RN) or recurrent brain tumor (rBT). Screening for rBT and RN at an early stage is crucial for facilitating faster treatment and better outcomes for the patients. Differentiating rBT from RN is challenging as both may present with similar radiological and clinical characteristics on MRI. Moreover, learning-based rBT versus RN classification using MRI may suffer from class imbalance due to lack of patient data. While synthetic data generation using generative models has shown promise to address class imbalance, the underlying data representation may be different in synthetic or augmented data. This study proposes computational modeling with statistically rigorous repeated random sub-sampling to balance the subset sample size for rBT and RN classification. The proposed pipeline includes multiresolution radiomic feature (MRF) extraction followed by feature selection with statistical significance testing (p<0.05). The five-fold cross validation results show the proposed model with MRF features classifies rBT from RN with an area under the curve (AUC) of 0.8920+-.055. Moreover, considering the dependence between survival time and censor time (where patients are not followed up until death), we demonstrate the feasibility of using MRF radiomic features as a non-invasive biomarker to identify patients who are at higher risk of recurrence or radiation necrosis. The cross-validated results show that the MRF model provides the best overall performance with an AUC of 0.770+-.032.
☆ Zero-Shot 3D Shape Correspondence
We propose a novel zero-shot approach to computing correspondences between 3D shapes. Existing approaches mainly focus on isometric and near-isometric shape pairs (e.g., human vs. human), but less attention has been given to strongly non-isometric and inter-class shape matching (e.g., human vs. cow). To this end, we introduce a fully automatic method that exploits the exceptional reasoning capabilities of recent foundation models in language and vision to tackle difficult shape correspondence problems. Our approach comprises multiple stages. First, we classify the 3D shapes in a zero-shot manner by feeding rendered shape views to a language-vision model (e.g., BLIP2) to generate a list of class proposals per shape. These proposals are unified into a single class per shape by employing the reasoning capabilities of ChatGPT. Second, we attempt to segment the two shapes in a zero-shot manner, but in contrast to the co-segmentation problem, we do not require a mutual set of semantic regions. Instead, we propose to exploit the in-context learning capabilities of ChatGPT to generate two different sets of semantic regions for each shape and a semantic mapping between them. This enables our approach to match strongly non-isometric shapes with significant differences in geometric structure. Finally, we employ the generated semantic mapping to produce coarse correspondences that can further be refined by the functional maps framework to produce dense point-to-point maps. Our approach, despite its simplicity, produces highly plausible results in a zero-shot manner, especially between strongly non-isometric shapes.
☆ Adversarial alignment: Breaking the trade-off between the strength of an attack and its relevance to human perception
Deep neural networks (DNNs) are known to have a fundamental sensitivity to adversarial attacks, perturbations of the input that are imperceptible to humans yet powerful enough to change the visual decision of a model. Adversarial attacks have long been considered the "Achilles' heel" of deep learning, which may eventually force a shift in modeling paradigms. Nevertheless, the formidable capabilities of modern large-scale DNNs have somewhat eclipsed these early concerns. Do adversarial attacks continue to pose a threat to DNNs? Here, we investigate how the robustness of DNNs to adversarial attacks has evolved as their accuracy on ImageNet has continued to improve. We measure adversarial robustness in two different ways: First, we measure the smallest adversarial attack needed to cause a model to change its object categorization decision. Second, we measure how aligned successful attacks are with the features that humans find diagnostic for object recognition. We find that adversarial attacks are inducing bigger and more easily detectable changes to image pixels as DNNs grow better on ImageNet, but these attacks are also becoming less aligned with features that humans find diagnostic for recognition. To better understand the source of this trade-off, we turn to the neural harmonizer, a DNN training routine that encourages models to leverage the same features as humans to solve tasks. Harmonized DNNs achieve the best of both worlds and experience attacks that are detectable and affect features that humans find diagnostic for recognition, meaning that attacks on these models are more likely to be rendered ineffective by inducing similar effects on human perception. Our findings suggest that the sensitivity of DNNs to adversarial attacks can be mitigated by DNN scale, data scale, and training routines that align models with biological intelligence.
☆ Discovering Novel Biological Traits From Images Using Phylogeny-Guided Neural Networks
Discovering evolutionary traits that are heritable across species on the tree of life (also referred to as a phylogenetic tree) is of great interest to biologists to understand how organisms diversify and evolve. However, the measurement of traits is often a subjective and labor-intensive process, making trait discovery a highly label-scarce problem. We present a novel approach for discovering evolutionary traits directly from images without relying on trait labels. Our proposed approach, Phylo-NN, encodes the image of an organism into a sequence of quantized feature vectors -- or codes -- where different segments of the sequence capture evolutionary signals at varying ancestry levels in the phylogeny. We demonstrate the effectiveness of our approach in producing biologically meaningful results in a number of downstream tasks including species image generation and species-to-species image translation, using fish species as a target example.
☆ Confidence-based federated distillation for vision-based lane-centering
A fundamental challenge of autonomous driving is maintaining the vehicle in the center of the lane by adjusting the steering angle. Recent advances leverage deep neural networks to predict steering decisions directly from images captured by the car cameras. Machine learning-based steering angle prediction needs to consider the vehicle's limitation in uploading large amounts of potentially private data for model training. Federated learning can address these constraints by enabling multiple vehicles to collaboratively train a global model without sharing their private data, but it is difficult to achieve good accuracy as the data distribution is often non-i.i.d. across the vehicles. This paper presents a new confidence-based federated distillation method to improve the performance of federated learning for steering angle prediction. Specifically, it proposes the novel use of entropy to determine the predictive confidence of each local model, and then selects the most confident local model as the teacher to guide the learning of the global model. A comprehensive evaluation of vision-based lane centering shows that the proposed approach can outperform FedAvg and FedDF by 11.3% and 9%, respectively.
comment: 5 pages, 5 figures
☆ MoDAR: Using Motion Forecasting for 3D Object Detection in Point Cloud Sequences CVPR 2023
Occluded and long-range objects are ubiquitous and challenging for 3D object detection. Point cloud sequence data provide unique opportunities to improve such cases, as an occluded or distant object can be observed from different viewpoints or gets better visibility over time. However, the efficiency and effectiveness in encoding long-term sequence data can still be improved. In this work, we propose MoDAR, using motion forecasting outputs as a type of virtual modality, to augment LiDAR point clouds. The MoDAR modality propagates object information from temporal contexts to a target frame, represented as a set of virtual points, one for each object from a waypoint on a forecasted trajectory. A fused point cloud of both raw sensor points and the virtual points can then be fed to any off-the-shelf point-cloud based 3D object detector. Evaluated on the Waymo Open Dataset, our method significantly improves prior art detectors by using motion forecasting from extra-long sequences (e.g. 18 seconds), achieving new state of the arts, while not adding much computation overhead.
comment: CVPR 2023
☆ ChatGPT as a mapping assistant: A novel method to enrich maps with generative AI and content derived from street-level photographs
This paper explores the concept of leveraging generative AI as a mapping assistant for enhancing the efficiency of collaborative mapping. We present results of an experiment that combines multiple sources of volunteered geographic information (VGI) and large language models (LLMs). Three analysts described the content of crowdsourced Mapillary street-level photographs taken along roads in a small test area in Miami, Florida. GPT-3.5-turbo was instructed to suggest the most appropriate tagging for each road in OpenStreetMap (OSM). The study also explores the utilization of BLIP-2, a state-of-the-art multimodal pre-training method as an artificial analyst of street-level photographs in addition to human analysts. Results demonstrate two ways to effectively increase the accuracy of mapping suggestions without modifying the underlying AI models: by (1) providing a more detailed description of source photographs, and (2) combining prompt engineering with additional context (e.g. location and objects detected along a road). The first approach increases the suggestion accuracy by up to 29%, and the second one by up to 20%.
comment: Submitted to The Fourth Spatial Data Science Symposium
☆ DeepVQE: Real Time Deep Voice Quality Enhancement for Joint Acoustic Echo Cancellation, Noise Suppression and Dereverberation
Acoustic echo cancellation (AEC), noise suppression (NS) and dereverberation (DR) are an integral part of modern full-duplex communication systems. As the demand for teleconferencing systems increases, addressing these tasks is required for an effective and efficient online meeting experience. Most prior research proposes solutions for these tasks separately, combining them with digital signal processing (DSP) based components, resulting in complex pipelines that are often impractical to deploy in real-world applications. This paper proposes a real-time cross-attention deep model, named DeepVQE, based on residual convolutional neural networks (CNNs) and recurrent neural networks (RNNs) to simultaneously address AEC, NS, and DR. We conduct several ablation studies to analyze the contributions of different components of our model to the overall performance. DeepVQE achieves state-of-the-art performance on non-personalized tracks from the ICASSP 2023 Acoustic Echo Cancellation Challenge and ICASSP 2023 Deep Noise Suppression Challenge test sets, showing that a single model can handle multiple tasks with excellent performance. Moreover, the model runs in real-time and has been successfully tested for the Microsoft Teams platform.
☆ Composition and Deformance: Measuring Imageability with a Text-to-Image Model
Although psycholinguists and psychologists have long studied the tendency of linguistic strings to evoke mental images in hearers or readers, most computational studies have applied this concept of imageability only to isolated words. Using recent developments in text-to-image generation models, such as DALLE mini, we propose computational methods that use generated images to measure the imageability of both single English words and connected text. We sample text prompts for image generation from three corpora: human-generated image captions, news article sentences, and poem lines. We subject these prompts to different deformances to examine the model's ability to detect changes in imageability caused by compositional change. We find high correlation between the proposed computational measures of imageability and human judgments of individual words. We also find the proposed measures more consistently respond to changes in compositionality than baseline approaches. We discuss possible effects of model training and implications for the study of compositionality in text-to-image models.
☆ DISCount: Counting in Large Image Collections with Detector-Based Importance Sampling
Many modern applications use computer vision to detect and count objects in massive image collections. However, when the detection task is very difficult or in the presence of domain shifts, the counts may be inaccurate even with significant investments in training data and model development. We propose DISCount -- a detector-based importance sampling framework for counting in large image collections that integrates an imperfect detector with human-in-the-loop screening to produce unbiased estimates of counts. We propose techniques for solving counting problems over multiple spatial or temporal regions using a small number of screened samples and estimate confidence intervals. This enables end-users to stop screening when estimates are sufficiently accurate, which is often the goal in a scientific study. On the technical side we develop variance reduction techniques based on control variates and prove the (conditional) unbiasedness of the estimators. DISCount leads to a 9-12x reduction in the labeling costs over naive screening for tasks we consider, such as counting birds in radar imagery or estimating damaged buildings in satellite imagery, and also surpasses alternative covariate-based screening approaches in efficiency.
♻ ☆ PaLI: A Jointly-Scaled Multilingual Language-Image Model ICLR 2023
Effective scaling and a flexible task interface enable large language models to excel at many tasks. We present PaLI (Pathways Language and Image model), a model that extends this approach to the joint modeling of language and vision. PaLI generates text based on visual and textual inputs, and with this interface performs many vision, language, and multimodal tasks, in many languages. To train PaLI, we make use of large pre-trained encoder-decoder language models and Vision Transformers (ViTs). This allows us to capitalize on their existing capabilities and leverage the substantial cost of training them. We find that joint scaling of the vision and language components is important. Since existing Transformers for language are much larger than their vision counterparts, we train a large, 4-billion parameter ViT (ViT-e) to quantify the benefits from even larger-capacity vision models. To train PaLI, we create a large multilingual mix of pretraining tasks, based on a new image-text training set containing 10B images and texts in over 100 languages. PaLI achieves state-of-the-art in multiple vision and language tasks (such as captioning, visual question-answering, scene-text understanding), while retaining a simple, modular, and scalable design.
comment: ICLR 2023 (Notable-top-5%)
♻ ☆ CLIP-Dissect: Automatic Description of Neuron Representations in Deep Vision Networks ICLR 2023
In this paper, we propose CLIP-Dissect, a new technique to automatically describe the function of individual hidden neurons inside vision networks. CLIP-Dissect leverages recent advances in multimodal vision/language models to label internal neurons with open-ended concepts without the need for any labeled data or human examples. We show that CLIP-Dissect provides more accurate descriptions than existing methods for last layer neurons where the ground-truth is available as well as qualitatively good descriptions for hidden layer neurons. In addition, our method is very flexible: it is model agnostic, can easily handle new concepts and can be extended to take advantage of better multimodal models in the future. Finally CLIP-Dissect is computationally efficient and can label all neurons from five layers of ResNet-50 in just 4 minutes, which is more than 10 times faster than existing methods. Our code is available at https://github.com/Trustworthy-ML-Lab/CLIP-dissect. Finally, crowdsourced user study results are available at Appendix B to further support the effectiveness of our method.
comment: Published in ICLR 2023 Conference (Spotlight). New v5(5 June 2023) - Added crowdsourced user study in Appendix B, not included in ICLR publication
♻ ☆ Robust 3D-aware Object Classification via Discriminative Render-and-Compare
In real-world applications, it is essential to jointly estimate the 3D object pose and class label of objects, i.e., to perform 3D-aware classification.While current approaches for either image classification or pose estimation can be extended to 3D-aware classification, we observe that they are inherently limited: 1) Their performance is much lower compared to the respective single-task models, and 2) they are not robust in out-of-distribution (OOD) scenarios. Our main contribution is a novel architecture for 3D-aware classification, which builds upon a recent work and performs comparably to single-task models while being highly robust. In our method, an object category is represented as a 3D cuboid mesh composed of feature vectors at each mesh vertex. Using differentiable rendering, we estimate the 3D object pose by minimizing the reconstruction error between the mesh and the feature representation of the target image. Object classification is then performed by comparing the reconstruction losses across object categories. Notably, the neural texture of the mesh is trained in a discriminative manner to enhance the classification performance while also avoiding local optima in the reconstruction loss. Furthermore, we show how our method and feed-forward neural networks can be combined to scale the render-and-compare approach to larger numbers of categories. Our experiments on PASCAL3D+, occluded-PASCAL3D+, and OOD-CV show that our method outperforms all baselines at 3D-aware classification by a wide margin in terms of performance and robustness.
♻ ☆ DSPDet3D: Dynamic Spatial Pruning for 3D Small Object Detection SP
Fine-grained 3D object detection is a core ability for agents to understand their 3D environment and interact with surrounding objects. However, current methods and benchmarks mainly focus on relatively large stuff. 3D object detectors still struggle on small objects due to weak geometric information. With in-depth study, we find increasing the spatial resolution of the feature maps significantly boosts the performance of 3D small object detection. And more interestingly, though the computational overhead increases dramatically with resolution, the growth mainly comes from the upsampling operation of the decoder. Inspired by this, we present a high-resolution multi-level detector with dynamic spatial pruning named DSPDet3D, which detects objects from large to small by iterative upsampling and meanwhile prunes the spatial representation of the scene at regions where there is no smaller object to be detected in higher resolution. We organize two benchmarks on ScanNet and TO-SCENE dataset to evaluate the ability of fine-grained 3D object detection, where our DSPDet3D improves the detection performance of small objects to a new level while achieving leading inference speed compared with existing 3D object detection methods. Moreover, DSPDet3D trained with only ScanNet rooms can generalize well to scenes in larger scale. It takes less than 2s for DSPDet3D to directly process a whole house or building consisting of dozens of rooms while detecting out almost all objects, ranging from bottles to beds, on a single RTX 3090 GPU. Project page: https://xuxw98.github.io/DSPDet3D/.
comment: Code is available at: https://github.com/xuxw98/DSPDet3D
♻ ☆ Label-Free Concept Bottleneck Models ICLR 2023
Concept bottleneck models (CBM) are a popular way of creating more interpretable neural networks by having hidden layer neurons correspond to human-understandable concepts. However, existing CBMs and their variants have two crucial limitations: first, they need to collect labeled data for each of the predefined concepts, which is time consuming and labor intensive; second, the accuracy of a CBM is often significantly lower than that of a standard neural network, especially on more complex datasets. This poor performance creates a barrier for adopting CBMs in practical real world applications. Motivated by these challenges, we propose Label-free CBM which is a novel framework to transform any neural network into an interpretable CBM without labeled concept data, while retaining a high accuracy. Our Label-free CBM has many advantages, it is: scalable - we present the first CBM scaled to ImageNet, efficient - creating a CBM takes only a few hours even for very large datasets, and automated - training it for a new dataset requires minimal human effort. Our code is available at https://github.com/Trustworthy-ML-Lab/Label-free-CBM. Finally, in Appendix B we conduct a large scale user evaluation of the interpretability of our method.
comment: Published at ICLR 2023. New v2(5 June 2023): added crowdsourced human study in Appendix B
♻ ☆ Reduce, Reuse, Recycle: Compositional Generation with Energy-Based Diffusion Models and MCMC ICML 2023
Since their introduction, diffusion models have quickly become the prevailing approach to generative modeling in many domains. They can be interpreted as learning the gradients of a time-varying sequence of log-probability density functions. This interpretation has motivated classifier-based and classifier-free guidance as methods for post-hoc control of diffusion models. In this work, we build upon these ideas using the score-based interpretation of diffusion models, and explore alternative ways to condition, modify, and reuse diffusion models for tasks involving compositional generation and guidance. In particular, we investigate why certain types of composition fail using current techniques and present a number of solutions. We conclude that the sampler (not the model) is responsible for this failure and propose new samplers, inspired by MCMC, which enable successful compositional generation. Further, we propose an energy-based parameterization of diffusion models which enables the use of new compositional operators and more sophisticated, Metropolis-corrected samplers. Intriguingly we find these samplers lead to notable improvements in compositional generation across a wide set of problems such as classifier-guided ImageNet modeling and compositional text-to-image generation.
comment: ICML 2023, Project Webpage: https://energy-based-model.github.io/reduce-reuse-recycle/
♻ ☆ Exploring the Connection between Robust and Generative Models
We offer a study that connects robust discriminative classifiers trained with adversarial training (AT) with generative modeling in the form of Energy-based Models (EBM). We do so by decomposing the loss of a discriminative classifier and showing that the discriminative model is also aware of the input data density. Though a common assumption is that adversarial points leave the manifold of the input data, our study finds out that, surprisingly, untargeted adversarial points in the input space are very likely under the generative model hidden inside the discriminative classifier -- have low energy in the EBM. We present two evidence: untargeted attacks are even more likely than the natural data and their likelihood increases as the attack strength increases. This allows us to easily detect them and craft a novel attack called High-Energy PGD that fools the classifier yet has energy similar to the data set. The code is available at github.com/senad96/Robust-Generative
comment: Italian Conference on AI - AI per Cybersecurity, 6 pages, 6 figures
♻ ☆ Linking generative semi-supervised learning and generative open-set recognition
This study investigates the relationship between semi-supervised learning (SSL) and open-set recognition (OSR) in the context of generative adversarial networks (GANs). Although no previous study has formally linked SSL and OSR, their respective methods share striking similarities. Specifically, SSL-GANs and OSR-GANs require their generators to produce samples in the complementary space. Subsequently, by regularising networks with generated samples, both SSL and OSR classifiers generalize the open space. To demonstrate the connection between SSL and OSR, we theoretically and experimentally compare state-of-the-art SSL-GAN methods with state-of-the-art OSR-GAN methods. Our results indicate that the SSL optimised margin-GANs, which have a stronger foundation in literature, set the new standard for the combined SSL-OSR task and achieves new state-of-other art results in certain general OSR experiments. However, the OSR optimised adversarial reciprocal point (ARP)-GANs still slightly out-performed margin-GANs at other OSR experiments. This result indicates unique insights for the combined optimisation task of SSL-OSR.
♻ ☆ Tutel: Adaptive Mixture-of-Experts at Scale
Sparsely-gated mixture-of-experts (MoE) has been widely adopted to scale deep learning models to trillion-plus parameters with fixed computational cost. The algorithmic performance of MoE relies on its token routing mechanism that forwards each input token to the right sub-models or experts. While token routing dynamically determines the amount of expert workload at runtime, existing systems suffer inefficient computation due to their static execution, namely static parallelism and pipelining, which does not adapt to the dynamic workload. We present Flex, a highly scalable stack design and implementation for MoE with dynamically adaptive parallelism and pipelining. Flex designs an identical layout for distributing MoE model parameters and input data, which can be leveraged by all possible parallelism or pipelining methods without any mathematical inequivalence or tensor migration overhead. This enables adaptive parallelism/pipelining optimization at zero cost during runtime. Based on this key design, Flex also implements various MoE acceleration techniques. Aggregating all techniques, Flex finally delivers huge speedup at any scale -- 4.96x and 5.75x speedup of a single MoE layer over 16 and 2,048 A100 GPUs, respectively, over the previous state-of-the-art. Our evaluation shows that Flex efficiently and effectively runs a real-world MoE-based model named SwinV2-MoE, built upon Swin Transformer V2, a state-of-the-art computer vision architecture. On efficiency, Flex accelerates SwinV2-MoE, achieving up to 1.55x and 2.11x speedup in training and inference over Fairseq, respectively. On effectiveness, the SwinV2-MoE model achieves superior accuracy in both pre-training and down-stream computer vision tasks such as COCO object detection than the counterpart dense model, indicating the readiness of Flex for end-to-end real-world model training and inference.
♻ ☆ MAMO: Masked Multimodal Modeling for Fine-Grained Vision-Language Representation Learning SIGIR 2023
Multimodal representation learning has shown promising improvements on various vision-language tasks. Most existing methods excel at building global-level alignment between vision and language while lacking effective fine-grained image-text interaction. In this paper, we propose a jointly masked multimodal modeling method to learn fine-grained multimodal representations. Our method performs joint masking on image-text input and integrates both implicit and explicit targets for the masked signals to recover. The implicit target provides a unified and debiased objective for vision and language, where the model predicts latent multimodal representations of the unmasked input. The explicit target further enriches the multimodal representations by recovering high-level and semantically meaningful information: momentum visual features of image patches and concepts of word tokens. Through such a masked modeling process, our model not only learns fine-grained multimodal interaction, but also avoids the semantic gap between high-level representations and low- or mid-level prediction targets (e.g. image pixels), thus producing semantically rich multimodal representations that perform well on both zero-shot and fine-tuned settings. Our pre-trained model (named MAMO) achieves state-of-the-art performance on various downstream vision-language tasks, including image-text retrieval, visual question answering, visual reasoning, and weakly-supervised visual grounding.
comment: SIGIR 2023, 10 pages
♻ ☆ Robust and Generalisable Segmentation of Subtle Epilepsy-causing Lesions: a Graph Convolutional Approach MICCAI 2023
Focal cortical dysplasia (FCD) is a leading cause of drug-resistant focal epilepsy, which can be cured by surgery. These lesions are extremely subtle and often missed even by expert neuroradiologists. "Ground truth" manual lesion masks are therefore expensive, limited and have large inter-rater variability. Existing FCD detection methods are limited by high numbers of false positive predictions, primarily due to vertex- or patch-based approaches that lack whole-brain context. Here, we propose to approach the problem as semantic segmentation using graph convolutional networks (GCN), which allows our model to learn spatial relationships between brain regions. To address the specific challenges of FCD identification, our proposed model includes an auxiliary loss to predict distance from the lesion to reduce false positives and a weak supervision classification loss to facilitate learning from uncertain lesion masks. On a multi-centre dataset of 1015 participants with surface-based features and manual lesion masks from structural MRI data, the proposed GCN achieved an AUC of 0.74, a significant improvement against a previously used vertex-wise multi-layer perceptron (MLP) classifier (AUC 0.64). With sensitivity thresholded at 67%, the GCN had a specificity of 71% in comparison to 49% when using the MLP. This improvement in specificity is vital for clinical integration of lesion-detection tools into the radiological workflow, through increasing clinical confidence in the use of AI radiological adjuncts and reducing the number of areas requiring expert review.
comment: accepted at MICCAI 2023
♻ ☆ Learning Prototype Classifiers for Long-Tailed Recognition IJCAI-23
The problem of long-tailed recognition (LTR) has received attention in recent years due to the fundamental power-law distribution of objects in the real-world. Most recent works in LTR use softmax classifiers that have a tendency to correlate classifier norm with the amount of training data for a given class. On the other hand, Prototype classifiers do not suffer from this shortcoming and can deliver promising results simply using Nearest-Class-Mean (NCM), a special case where prototypes are empirical centroids. However, the potential of Prototype classifiers as an alternative to softmax in LTR is relatively underexplored. In this work, we propose Prototype classifiers, which jointly learn prototypes that minimize average cross-entropy loss based on probability scores from distances to prototypes. We theoretically analyze the properties of Euclidean distance based prototype classifiers that leads to stable gradient-based optimization which is robust to outliers. We further enhance Prototype classifiers by learning channel-dependent temperature parameters to enable independent distance scales along each channel. Our analysis shows that prototypes learned by Prototype classifiers are better separated than empirical centroids. Results on four long-tailed recognition benchmarks show that Prototype classifier outperforms or is comparable to the state-of-the-art methods.
comment: Accepted at IJCAI-23
♻ ☆ RLSbench: Domain Adaptation Under Relaxed Label Shift ICML 2023
Despite the emergence of principled methods for domain adaptation under label shift, their sensitivity to shifts in class conditional distributions is precariously under explored. Meanwhile, popular deep domain adaptation heuristics tend to falter when faced with label proportions shifts. While several papers modify these heuristics in attempts to handle label proportions shifts, inconsistencies in evaluation standards, datasets, and baselines make it difficult to gauge the current best practices. In this paper, we introduce RLSbench, a large-scale benchmark for relaxed label shift, consisting of $>$500 distribution shift pairs spanning vision, tabular, and language modalities, with varying label proportions. Unlike existing benchmarks, which primarily focus on shifts in class-conditional $p(x|y)$, our benchmark also focuses on label marginal shifts. First, we assess 13 popular domain adaptation methods, demonstrating more widespread failures under label proportion shifts than were previously known. Next, we develop an effective two-step meta-algorithm that is compatible with most domain adaptation heuristics: (i) pseudo-balance the data at each epoch; and (ii) adjust the final classifier with target label distribution estimate. The meta-algorithm improves existing domain adaptation heuristics under large label proportion shifts, often by 2--10\% accuracy points, while conferring minimal effect ($<$0.5\%) when label proportions do not shift. We hope that these findings and the availability of RLSbench will encourage researchers to rigorously evaluate proposed methods in relaxed label shift settings. Code is publicly available at https://github.com/acmi-lab/RLSbench.
comment: Accepted at ICML 2023. Paper website: https://sites.google.com/view/rlsbench/
♻ ☆ Evaluation of Interpretability Methods and Perturbation Artifacts in Deep Neural Networks
Despite excellent performance of deep neural networks (DNNs) in image classification, detection, and prediction, characterizing how DNNs make a given decision remains an open problem, resulting in a number of interpretability methods. Post-hoc interpretability methods primarily aim to quantify the importance of input features with respect to the class probabilities. However, due to the lack of ground truth and the existence of interpretability methods with diverse operating characteristics, evaluating these methods is a crucial challenge. A popular approach to evaluate interpretability methods is to perturb input features deemed important for a given prediction and observe the decrease in accuracy. However, perturbation itself may introduce artifacts. We propose a method for estimating the impact of such artifacts on the fidelity estimation by utilizing model accuracy curves from perturbing input features according to the Most Import First (MIF) and Least Import First (LIF) orders. Using the ResNet-50 trained on the ImageNet, we demonstrate the proposed fidelity estimation of four popular post-hoc interpretability methods.
comment: 11 pages, 5 figures
♻ ☆ Signs of Language: Embodied Sign Language Fingerspelling Acquisition from Demonstrations for Human-Robot Interaction
Learning fine-grained movements is a challenging topic in robotics, particularly in the context of robotic hands. One specific instance of this challenge is the acquisition of fingerspelling sign language in robots. In this paper, we propose an approach for learning dexterous motor imitation from video examples without additional information. To achieve this, we first build a URDF model of a robotic hand with a single actuator for each joint. We then leverage pre-trained deep vision models to extract the 3D pose of the hand from RGB videos. Next, using state-of-the-art reinforcement learning algorithms for motion imitation (namely, proximal policy optimization and soft actor-critic), we train a policy to reproduce the movement extracted from the demonstrations. We identify the optimal set of hyperparameters for imitation based on a reference motion. Finally, we demonstrate the generalizability of our approach by testing it on six different tasks, corresponding to fingerspelled letters. Our results show that our approach is able to successfully imitate these fine-grained movements without additional information, highlighting its potential for real-world applications in robotics.
♻ ☆ Considerations on the Evaluation of Biometric Quality Assessment Algorithms
Quality assessment algorithms can be used to estimate the utility of a biometric sample for the purpose of biometric recognition. "Error versus Discard Characteristic" (EDC) plots, and "partial Area Under Curve" (pAUC) values of curves therein, are generally used by researchers to evaluate the predictive performance of such quality assessment algorithms. An EDC curve depends on an error type such as the "False Non Match Rate" (FNMR), a quality assessment algorithm, a biometric recognition system, a set of comparisons each corresponding to a biometric sample pair, and a comparison score threshold corresponding to a starting error. To compute an EDC curve, comparisons are progressively discarded based on the associated samples' lowest quality scores, and the error is computed for the remaining comparisons. Additionally, a discard fraction limit or range must be selected to compute pAUC values, which can then be used to quantitatively rank quality assessment algorithms. This paper discusses and analyses various details for this kind of quality assessment algorithm evaluation, including general EDC properties, interpretability improvements for pAUC values based on a hard lower error limit and a soft upper error limit, the use of relative instead of discrete rankings, stepwise vs. linear curve interpolation, and normalisation of quality scores to a [0, 100] integer range. We also analyse the stability of quantitative quality assessment algorithm rankings based on pAUC values across varying pAUC discard fraction limits and starting errors, concluding that higher pAUC discard fraction limits should be preferred. The analyses are conducted both with synthetic data and with real data for a face image quality assessment scenario, with a focus on general modality-independent conclusions for EDC evaluations.
♻ ☆ Construction of Hierarchical Neural Architecture Search Spaces based on Context-free Grammars
The discovery of neural architectures from simple building blocks is a long-standing goal of Neural Architecture Search (NAS). Hierarchical search spaces are a promising step towards this goal but lack a unifying search space design framework and typically only search over some limited aspect of architectures. In this work, we introduce a unifying search space design framework based on context-free grammars that can naturally and compactly generate expressive hierarchical search spaces that are 100s of orders of magnitude larger than common spaces from the literature. By enhancing and using their properties, we effectively enable search over the complete architecture and can foster regularity. Further, we propose an efficient hierarchical kernel design for a Bayesian Optimization search strategy to efficiently search over such huge spaces. We demonstrate the versatility of our search space design framework and show that our search strategy can be superior to existing NAS approaches. Code is available at https://github.com/automl/hierarchical_nas_construction.
♻ ☆ Learning the Relation between Similarity Loss and Clustering Loss in Self-Supervised Learning
Self-supervised learning enables networks to learn discriminative features from massive data itself. Most state-of-the-art methods maximize the similarity between two augmentations of one image based on contrastive learning. By utilizing the consistency of two augmentations, the burden of manual annotations can be freed. Contrastive learning exploits instance-level information to learn robust features. However, the learned information is probably confined to different views of the same instance. In this paper, we attempt to leverage the similarity between two distinct images to boost representation in self-supervised learning. In contrast to instance-level information, the similarity between two distinct images may provide more useful information. Besides, we analyze the relation between similarity loss and feature-level cross-entropy loss. These two losses are essential for most deep learning methods. However, the relation between these two losses is not clear. Similarity loss helps obtain instance-level representation, while feature-level cross-entropy loss helps mine the similarity between two distinct images. We provide theoretical analyses and experiments to show that a suitable combination of these two losses can get state-of-the-art results. Code is available at https://github.com/guijiejie/ICCL.
comment: This paper is accepted by IEEE Transactions on Image Processing
♻ ☆ Generalist Vision Foundation Models for Medical Imaging: A Case Study of Segment Anything Model on Zero-Shot Medical Segmentation
In this paper, we examine the recent Segment Anything Model (SAM) on medical images, and report both quantitative and qualitative zero-shot segmentation results on nine medical image segmentation benchmarks, covering various imaging modalities, such as optical coherence tomography (OCT), magnetic resonance imaging (MRI), and computed tomography (CT), as well as different applications including dermatology, ophthalmology, and radiology. Those benchmarks are representative and commonly used in model development. Our experimental results indicate that while SAM presents remarkable segmentation performance on images from the general domain, its zero-shot segmentation ability remains restricted for out-of-distribution images, e.g., medical images. In addition, SAM exhibits inconsistent zero-shot segmentation performance across different unseen medical domains. For certain structured targets, e.g., blood vessels, the zero-shot segmentation of SAM completely failed. In contrast, a simple fine-tuning of it with a small amount of data could lead to remarkable improvement of the segmentation quality, showing the great potential and feasibility of using fine-tuned SAM to achieve accurate medical image segmentation for a precision diagnostics. Our study indicates the versatility of generalist vision foundation models on medical imaging, and their great potential to achieve desired performance through fine-turning and eventually address the challenges associated with accessing large and diverse medical datasets in support of clinical diagnostics.
comment: Published in Diagnostics
♻ ☆ Candidate Set Re-ranking for Composed Image Retrieval with Dual Multi-modal Encoder
Composed image retrieval aims to find an image that best matches a given multi-modal user query consisting of a reference image and text pair. Existing methods commonly pre-compute image embeddings over the entire corpus and compare these to a reference image embedding modified by the query text at test time. Such a pipeline is very efficient at test time since fast vector distances can be used to evaluate candidates, but modifying the reference image embedding guided only by a short textual description can be difficult, especially independent of potential candidates. An alternative approach is to allow interactions between the query and every possible candidate, i.e., reference-text-candidate triplets, and pick the best from the entire set. Though this approach is more discriminative, for large-scale datasets the computational cost is prohibitive since pre-computation of candidate embeddings is no longer possible. We propose to combine the merits of both schemes using a two-stage model. Our first stage adopts the conventional vector distancing metric and performs a fast pruning among candidates. Meanwhile, our second stage employs a dual-encoder architecture, which effectively attends to the input triplet of reference-text-candidate and re-ranks the candidates. Both stages utilize a vision-and-language pre-trained network, which has proven beneficial for various downstream tasks. Our method consistently outperforms state-of-the-art approaches on standard benchmarks for the task.
comment: 14 pages
♻ ☆ Diffusion-Based Hierarchical Multi-Label Object Detection to Analyze Panoramic Dental X-rays MICCAI 2023
Due to the necessity for precise treatment planning, the use of panoramic X-rays to identify different dental diseases has tremendously increased. Although numerous ML models have been developed for the interpretation of panoramic X-rays, there has not been an end-to-end model developed that can identify problematic teeth with dental enumeration and associated diagnoses at the same time. To develop such a model, we structure the three distinct types of annotated data hierarchically following the FDI system, the first labeled with only quadrant, the second labeled with quadrant-enumeration, and the third fully labeled with quadrant-enumeration-diagnosis. To learn from all three hierarchies jointly, we introduce a novel diffusion-based hierarchical multi-label object detection framework by adapting a diffusion-based method that formulates object detection as a denoising diffusion process from noisy boxes to object boxes. Specifically, to take advantage of the hierarchically annotated data, our method utilizes a novel noisy box manipulation technique by adapting the denoising process in the diffusion network with the inference from the previously trained model in hierarchical order. We also utilize a multi-label object detection method to learn efficiently from partial annotations and to give all the needed information about each abnormal tooth for treatment planning. Experimental results show that our method significantly outperforms state-of-the-art object detection methods, including RetinaNet, Faster R-CNN, DETR, and DiffusionDet for the analysis of panoramic X-rays, demonstrating the great potential of our method for hierarchically and partially annotated datasets. The code and the data are available at: https://github.com/ibrahimethemhamamci/HierarchicalDet.
comment: MICCAI 2023
♻ ☆ ChatFace: Chat-Guided Real Face Editing via Diffusion Latent Space Manipulation
Editing real facial images is a crucial task in computer vision with significant demand in various real-world applications. While GAN-based methods have showed potential in manipulating images especially when combined with CLIP, these methods are limited in their ability to reconstruct real images due to challenging GAN inversion capability. Despite the successful image reconstruction achieved by diffusion-based methods, there are still challenges in effectively manipulating fine-gained facial attributes with textual instructions.To address these issues and facilitate convenient manipulation of real facial images, we propose a novel approach that conduct text-driven image editing in the semantic latent space of diffusion model. By aligning the temporal feature of the diffusion model with the semantic condition at generative process, we introduce a stable manipulation strategy, which perform precise zero-shot manipulation effectively. Furthermore, we develop an interactive system named ChatFace, which combines the zero-shot reasoning ability of large language models to perform efficient manipulations in diffusion semantic latent space. This system enables users to perform complex multi-attribute manipulations through dialogue, opening up new possibilities for interactive image editing. Extensive experiments confirmed that our approach outperforms previous methods and enables precise editing of real facial images, making it a promising candidate for real-world applications. Project page: https://dongxuyue.github.io/chatface/
♻ ☆ Learning Point-Language Hierarchical Alignment for 3D Visual Grounding ECCV 2022
This paper presents a novel hierarchical alignment model (HAM) that learns multi-granularity visual and linguistic representations in an end-to-end manner. We extract key points and proposal points to model 3D contexts and instances, and propose point-language alignment with context modulation (PLACM) mechanism, which learns to gradually align word-level and sentence-level linguistic embeddings with visual representations, while the modulation with the visual context captures latent informative relationships. To further capture both global and local relationships, we propose a spatially multi-granular modeling scheme that applies PLACM to both global and local fields. Experimental results demonstrate the superiority of HAM, with visualized results showing that it can dynamically model fine-grained visual and linguistic representations. HAM outperforms existing methods by a significant margin and achieves state-of-the-art performance on two publicly available datasets, and won the championship in ECCV 2022 ScanRefer challenge. Code is available at~\url{https://github.com/PPjmchen/HAM}.
comment: Champion on ECCV 2022 ScanRefer Challenge
♻ ☆ Fully Context-Aware Image Inpainting with a Learned Semantic Pyramid
Restoring reasonable and realistic content for arbitrary missing regions in images is an important yet challenging task. Although recent image inpainting models have made significant progress in generating vivid visual details, they can still lead to texture blurring or structural distortions due to contextual ambiguity when dealing with more complex scenes. To address this issue, we propose the Semantic Pyramid Network (SPN) motivated by the idea that learning multi-scale semantic priors from specific pretext tasks can greatly benefit the recovery of locally missing content in images. SPN consists of two components. First, it distills semantic priors from a pretext model into a multi-scale feature pyramid, achieving a consistent understanding of the global context and local structures. Within the prior learner, we present an optional module for variational inference to realize probabilistic image inpainting driven by various learned priors. The second component of SPN is a fully context-aware image generator, which adaptively and progressively refines low-level visual representations at multiple scales with the (stochastic) prior pyramid. We train the prior learner and the image generator as a unified model without any post-processing. Our approach achieves the state of the art on multiple datasets, including Places2, Paris StreetView, CelebA, and CelebA-HQ, under both deterministic and probabilistic inpainting setups.
comment: Accepted by Pattern Recognition, 2023
♻ ☆ Image Reconstruction via Deep Image Prior Subspaces
Deep learning has been widely used for solving image reconstruction tasks but its deployability has been held back due to the shortage of high-quality training data. Unsupervised learning methods, such as the deep image prior (DIP), naturally fill this gap, but bring a host of new issues: the susceptibility to overfitting due to a lack of robust early stopping strategies and unstable convergence. We present a novel approach to tackle these issues by restricting DIP optimisation to a sparse linear subspace of its parameters, employing a synergy of dimensionality reduction techniques and second order optimisation methods. The low-dimensionality of the subspace reduces DIP's tendency to fit noise and allows the use of stable second order optimisation methods, e.g., natural gradient descent or L-BFGS. Experiments across both image restoration and tomographic tasks of different geometry and ill-posedness show that second order optimisation within a low-dimensional subspace is favourable in terms of optimisation stability to reconstruction fidelity trade-off.
♻ ☆ ImplantFormer: Vision Transformer based Implant Position Regression Using Dental CBCT Data
Implant prosthesis is the most appropriate treatment for dentition defect or dentition loss, which usually involves a surgical guide design process to decide the implant position. However, such design heavily relies on the subjective experiences of dentists. In this paper, a transformer-based Implant Position Regression Network, ImplantFormer, is proposed to automatically predict the implant position based on the oral CBCT data. We creatively propose to predict the implant position using the 2D axial view of the tooth crown area and fit a centerline of the implant to obtain the actual implant position at the tooth root. Convolutional stem and decoder are designed to coarsely extract image features before the operation of patch embedding and integrate multi-level feature maps for robust prediction, respectively. As both long-range relationship and local features are involved, our approach can better represent global information and achieves better location performance. Extensive experiments on a dental implant dataset through five-fold cross-validation demonstrated that the proposed ImplantFormer achieves superior performance than existing methods.
♻ ☆ Introducing Depth into Transformer-based 3D Object Detection
In this paper, we present DAT, a Depth-Aware Transformer framework designed for camera-based 3D detection. Our model is based on observing two major issues in existing methods: large depth translation errors and duplicate predictions along depth axes. To mitigate these issues, we propose two key solutions within DAT. To address the first issue, we introduce a Depth-Aware Spatial Cross-Attention (DA-SCA) module that incorporates depth information into spatial cross-attention when lifting image features to 3D space. To address the second issue, we introduce an auxiliary learning task called Depth-aware Negative Suppression loss. First, based on their reference points, we organize features as a Bird's-Eye-View (BEV) feature map. Then, we sample positive and negative features along each object ray that connects an object and a camera and train the model to distinguish between them. The proposed DA-SCA and DNS methods effectively alleviate these two problems. We show that DAT is a versatile method that enhances the performance of all three popular models, BEVFormer, DETR3D, and PETR. Our evaluation on BEVFormer demonstrates that DAT achieves a significant improvement of +2.8 NDS on nuScenes val under the same settings. Moreover, when using pre-trained VoVNet-99 as the backbone, DAT achieves strong results of 60.0 NDS and 51.5 mAP on nuScenes test. Our code will be soon.
comment: revision
♻ ☆ Discover and Cure: Concept-aware Mitigation of Spurious Correlation ICML 2023
Deep neural networks often rely on spurious correlations to make predictions, which hinders generalization beyond training environments. For instance, models that associate cats with bed backgrounds can fail to predict the existence of cats in other environments without beds. Mitigating spurious correlations is crucial in building trustworthy models. However, the existing works lack transparency to offer insights into the mitigation process. In this work, we propose an interpretable framework, Discover and Cure (DISC), to tackle the issue. With human-interpretable concepts, DISC iteratively 1) discovers unstable concepts across different environments as spurious attributes, then 2) intervenes on the training data using the discovered concepts to reduce spurious correlation. Across systematic experiments, DISC provides superior generalization ability and interpretability than the existing approaches. Specifically, it outperforms the state-of-the-art methods on an object recognition task and a skin-lesion classification task by 7.5% and 9.6%, respectively. Additionally, we offer theoretical analysis and guarantees to understand the benefits of models trained by DISC. Code and data are available at https://github.com/Wuyxin/DISC.
comment: ICML 2023
♻ ☆ ClusterFuG: Clustering Fully connected Graphs by Multicut ICML 2023
We propose a graph clustering formulation based on multicut (a.k.a. weighted correlation clustering) on the complete graph. Our formulation does not need specification of the graph topology as in the original sparse formulation of multicut, making our approach simpler and potentially better performing. In contrast to unweighted correlation clustering we allow for a more expressive weighted cost structure. In dense multicut, the clustering objective is given in a factorized form as inner products of node feature vectors. This allows for an efficient formulation and inference in contrast to multicut/weighted correlation clustering, which has at least quadratic representation and computation complexity when working on the complete graph. We show how to rewrite classical greedy algorithms for multicut in our dense setting and how to modify them for greater efficiency and solution quality. In particular, our algorithms scale to graphs with tens of thousands of nodes. Empirical evidence on instance segmentation on Cityscapes and clustering of ImageNet datasets shows the merits of our approach.
comment: ICML 2023
♻ ☆ SelFLoc: Selective Feature Fusion for Large-scale Point Cloud-based Place Recognition
Point cloud-based place recognition is crucial for mobile robots and autonomous vehicles, especially when the global positioning sensor is not accessible. LiDAR points are scattered on the surface of objects and buildings, which have strong shape priors along different axes. To enhance message passing along particular axes, Stacked Asymmetric Convolution Block (SACB) is designed, which is one of the main contributions in this paper. Comprehensive experiments demonstrate that asymmetric convolution and its corresponding strategies employed by SACB can contribute to the more effective representation of point cloud feature. On this basis, Selective Feature Fusion Block (SFFB), which is formed by stacking point- and channel-wise gating layers in a predefined sequence, is proposed to selectively boost salient local features in certain key regions, as well as to align the features before fusion phase. SACBs and SFFBs are combined to construct a robust and accurate architecture for point cloud-based place recognition, which is termed SelFLoc. Comparative experimental results show that SelFLoc achieves the state-of-the-art (SOTA) performance on the Oxford and other three in-house benchmarks with an improvement of 1.6 absolute percentages on mean average recall@1.
♻ ☆ FSD: Fully-Specialized Detector via Neural Architecture Search
Most generic object detectors are mainly built for standard object detection tasks such as COCO and PASCAL VOC. They might not work well and/or efficiently on tasks of other domains consisting of images that are visually different from standard datasets. To this end, many advances have been focused on adapting a general-purposed object detector with limited domain-specific designs. However, designing a successful task-specific detector requires extraneous manual experiments and parameter tuning through trial and error. In this paper, we first propose and examine a fully-automatic pipeline to design a fully-specialized detector (FSD) which mainly incorporates a neural-architectural-searched model by exploring ideal network structures over the backbone and task-specific head. On the DeepLesion dataset, extensive results show that FSD can achieve 3.1 mAP gain while using approximately 40% fewer parameters on binary lesion detection task and improved the mAP by around 10% on multi-type lesion detection task via our region-aware graph modeling compared with existing general-purposed medical lesion detection networks.
♻ ☆ Transferable Adversarial Attacks on Vision Transformers with Token Gradient Regularization CVPR 2023
Vision transformers (ViTs) have been successfully deployed in a variety of computer vision tasks, but they are still vulnerable to adversarial samples. Transfer-based attacks use a local model to generate adversarial samples and directly transfer them to attack a target black-box model. The high efficiency of transfer-based attacks makes it a severe security threat to ViT-based applications. Therefore, it is vital to design effective transfer-based attacks to identify the deficiencies of ViTs beforehand in security-sensitive scenarios. Existing efforts generally focus on regularizing the input gradients to stabilize the updated direction of adversarial samples. However, the variance of the back-propagated gradients in intermediate blocks of ViTs may still be large, which may make the generated adversarial samples focus on some model-specific features and get stuck in poor local optima. To overcome the shortcomings of existing approaches, we propose the Token Gradient Regularization (TGR) method. According to the structural characteristics of ViTs, TGR reduces the variance of the back-propagated gradient in each internal block of ViTs in a token-wise manner and utilizes the regularized gradient to generate adversarial samples. Extensive experiments on attacking both ViTs and CNNs confirm the superiority of our approach. Notably, compared to the state-of-the-art transfer-based attacks, our TGR offers a performance improvement of 8.8% on average.
comment: CVPR 2023, Code is available at https://github.com/jpzhang1810/TGR
♻ ☆ Learning Object-Centric Neural Scattering Functions for Free-viewpoint Relighting and Scene Composition
Photorealistic object appearance modeling from 2D images is a constant topic in vision and graphics. While neural implicit methods (such as Neural Radiance Fields) have shown high-fidelity view synthesis results, they cannot relight the captured objects. More recent neural inverse rendering approaches have enabled object relighting, but they represent surface properties as simple BRDFs, and therefore cannot handle translucent objects. We propose Object-Centric Neural Scattering Functions (OSFs) for learning to reconstruct object appearance from only images. OSFs not only support free-viewpoint object relighting, but also can model both opaque and translucent objects. While accurately modeling subsurface light transport for translucent objects can be highly complex and even intractable for neural methods, OSFs learn to approximate the radiance transfer from a distant light to an outgoing direction at any spatial location. This approximation avoids explicitly modeling complex subsurface scattering, making learning a neural implicit model tractable. Experiments on real and synthetic data show that OSFs accurately reconstruct appearances for both opaque and translucent objects, allowing faithful free-viewpoint relighting as well as scene composition. Project website: https://kovenyu.com/osf/
comment: Project website: https://kovenyu.com/osf/ Journal extension of arXiv:2012.08503. The first two authors contributed equally to this work
♻ ☆ Object as Query: Lifting any 2D Object Detector to 3D Detection
3D object detection from multi-view images has drawn much attention over the past few years. Existing methods mainly establish 3D representations from multi-view images and adopt a dense detection head for object detection, or employ object queries distributed in 3D space to localize objects. In this paper, we design Multi-View 2D Objects guided 3D Object Detector (MV2D), which can lift any 2D object detector to multi-view 3D object detection. Since 2D detections can provide valuable priors for object existence, MV2D exploits 2D detectors to generate object queries conditioned on the rich image semantics. These dynamically generated queries help MV2D to recall objects in the field of view and show a strong capability of localizing 3D objects. For the generated queries, we design a sparse cross attention module to force them to focus on the features of specific objects, which suppresses interference from noises. The evaluation results on the nuScenes dataset demonstrate the dynamic object queries and sparse feature aggregation can promote 3D detection capability. MV2D also exhibits a state-of-the-art performance among existing methods. We hope MV2D can serve as a new baseline for future research.
comment: technical report
♻ ☆ Q2ATransformer: Improving Medical VQA via an Answer Querying Decoder
Medical Visual Question Answering (VQA) systems play a supporting role to understand clinic-relevant information carried by medical images. The questions to a medical image include two categories: close-end (such as Yes/No question) and open-end. To obtain answers, the majority of the existing medical VQA methods relies on classification approaches, while a few works attempt to use generation approaches or a mixture of the two. The classification approaches are relatively simple but perform poorly on long open-end questions. To bridge this gap, in this paper, we propose a new Transformer based framework for medical VQA (named as Q2ATransformer), which integrates the advantages of both the classification and the generation approaches and provides a unified treatment for the close-end and open-end questions. Specifically, we introduce an additional Transformer decoder with a set of learnable candidate answer embeddings to query the existence of each answer class to a given image-question pair. Through the Transformer attention, the candidate answer embeddings interact with the fused features of the image-question pair to make the decision. In this way, despite being a classification-based approach, our method provides a mechanism to interact with the answer information for prediction like the generation-based approaches. On the other hand, by classification, we mitigate the task difficulty by reducing the search space of answers. Our method achieves new state-of-the-art performance on two medical VQA benchmarks. Especially, for the open-end questions, we achieve 79.19% on VQA-RAD and 54.85% on PathVQA, with 16.09% and 41.45% absolute improvements, respectively.
♻ ☆ Spatio-Temporal AU Relational Graph Representation Learning For Facial Action Units Detection
This paper presents our Facial Action Units (AUs) detection submission to the fifth Affective Behavior Analysis in-the-wild Competition (ABAW). Our approach consists of three main modules: (i) a pre-trained facial representation encoder which produce a strong facial representation from each input face image in the input sequence; (ii) an AU-specific feature generator that specifically learns a set of AU features from each facial representation; and (iii) a spatio-temporal graph learning module that constructs a spatio-temporal graph representation. This graph representation describes AUs contained in all frames and predicts the occurrence of each AU based on both the modeled spatial information within the corresponding face and the learned temporal dynamics among frames. The experimental results show that our approach outperformed the baseline and the spatio-temporal graph representation learning allows our model to generate the best results among all ablated systems. Our model ranks at the 4th place in the AU recognition track at the 5th ABAW Competition. Our code is publicly available at https://github.com/wzh125/ABAW-5.
♻ ☆ A Novel Vision Transformer with Residual in Self-attention for Biomedical Image Classification
Biomedical image classification requires capturing of bio-informatics based on specific feature distribution. In most of such applications, there are mainly challenges due to limited availability of samples for diseased cases and imbalanced nature of dataset. This article presents the novel framework of multi-head self-attention for vision transformer (ViT) which makes capable of capturing the specific image features for classification and analysis. The proposed method uses the concept of residual connection for accumulating the best attention output in each block of multi-head attention. The proposed framework has been evaluated on two small datasets: (i) blood cell classification dataset and (ii) brain tumor detection using brain MRI images. The results show the significant improvement over traditional ViT and other convolution based state-of-the-art classification models.
♻ ☆ PanoContext-Former: Panoramic Total Scene Understanding with a Transformer
Panoramic image enables deeper understanding and more holistic perception of $360^\circ$ surrounding environment, which can naturally encode enriched scene context information compared to standard perspective image. Previous work has made lots of effort to solve the scene understanding task in a bottom-up form, thus each sub-task is processed separately and few correlations are explored in this procedure. In this paper, we propose a novel method using depth prior for holistic indoor scene understanding which recovers the objects' shapes, oriented bounding boxes and the 3D room layout simultaneously from a single panorama. In order to fully utilize the rich context information, we design a transformer-based context module to predict the representation and relationship among each component of the scene. In addition, we introduce a real-world dataset for scene understanding, including photo-realistic panoramas, high-fidelity depth images, accurately annotated room layouts, and oriented object bounding boxes and shapes. Experiments on the synthetic and real-world datasets demonstrate that our method outperforms previous panoramic scene understanding methods in terms of both layout estimation and 3D object detection.
♻ ☆ X-Pruner: eXplainable Pruning for Vision Transformers
Recently vision transformer models have become prominent models for a range of tasks. These models, however, usually suffer from intensive computational costs and heavy memory requirements, making them impractical for deployment on edge platforms. Recent studies have proposed to prune transformers in an unexplainable manner, which overlook the relationship between internal units of the model and the target class, thereby leading to inferior performance. To alleviate this problem, we propose a novel explainable pruning framework dubbed X-Pruner, which is designed by considering the explainability of the pruning criterion. Specifically, to measure each prunable unit's contribution to predicting each target class, a novel explainability-aware mask is proposed and learned in an end-to-end manner. Then, to preserve the most informative units and learn the layer-wise pruning rate, we adaptively search the layer-wise threshold that differentiates between unpruned and pruned units based on their explainability-aware mask values. To verify and evaluate our method, we apply the X-Pruner on representative transformer models including the DeiT and Swin Transformer. Comprehensive simulation results demonstrate that the proposed X-Pruner outperforms the state-of-the-art black-box methods with significantly reduced computational costs and slight performance degradation.
♻ ☆ DC-Net: Divide-and-Conquer for Salient Object Detection
In this paper, we introduce Divide-and-Conquer into the salient object detection (SOD) task to enable the model to learn prior knowledge that is for predicting the saliency map. We design a novel network, Divide-and-Conquer Network (DC-Net) which uses two encoders to solve different subtasks that are conducive to predicting the final saliency map, here is to predict the edge maps with width 4 and location maps of salient objects and then aggregate the feature maps with different semantic information into the decoder to predict the final saliency map. The decoder of DC-Net consists of our newly designed two-level Residual nested-ASPP (ResASPP$^{2}$) modules, which have the ability to capture a large number of different scale features with a small number of convolution operations and have the advantages of maintaining high resolution all the time and being able to obtain a large and compact effective receptive field (ERF). Based on the advantage of Divide-and-Conquer's parallel computing, we use Parallel Acceleration to speed up DC-Net, allowing it to achieve competitive performance on six LR-SOD and five HR-SOD datasets under high efficiency (60 FPS and 55 FPS). Codes and results are available: https://github.com/PiggyJerry/DC-Net.
♻ ☆ LiDAR2Map: In Defense of LiDAR-Based Semantic Map Construction Using Online Camera Distillation CVPR2023
Semantic map construction under bird's-eye view (BEV) plays an essential role in autonomous driving. In contrast to camera image, LiDAR provides the accurate 3D observations to project the captured 3D features onto BEV space inherently. However, the vanilla LiDAR-based BEV feature often contains many indefinite noises, where the spatial features have little texture and semantic cues. In this paper, we propose an effective LiDAR-based method to build semantic map. Specifically, we introduce a BEV feature pyramid decoder that learns the robust multi-scale BEV features for semantic map construction, which greatly boosts the accuracy of the LiDAR-based method. To mitigate the defects caused by lacking semantic cues in LiDAR data, we present an online Camera-to-LiDAR distillation scheme to facilitate the semantic learning from image to point cloud. Our distillation scheme consists of feature-level and logit-level distillation to absorb the semantic information from camera in BEV. The experimental results on challenging nuScenes dataset demonstrate the efficacy of our proposed LiDAR2Map on semantic map construction, which significantly outperforms the previous LiDAR-based methods over 27.9% mIoU and even performs better than the state-of-the-art camera-based approaches. Source code is available at: https://github.com/songw-zju/LiDAR2Map.
comment: Accepted by CVPR2023
♻ ☆ Joint fMRI Decoding and Encoding with Latent Embedding Alignment
The connection between brain activity and corresponding visual stimuli is crucial in comprehending the human brain. While deep generative models have exhibited advancement in recovering brain recordings by generating images conditioned on fMRI signals, accomplishing high-quality generation with consistent semantics continues to pose challenges. Moreover, the prediction of brain activity from visual stimuli remains a formidable undertaking. In this paper, we introduce a unified framework that addresses both fMRI decoding and encoding. Commencing with the establishment of two latent spaces capable of representing and reconstructing fMRI signals and visual images, respectively, we proceed to align the fMRI signals and visual images within the latent space, thereby enabling a bidirectional transformation between the two domains. Our Latent Embedding Alignment (LEA) model concurrently recovers visual stimuli from fMRI signals and predicts brain activity from images within a unified framework. The performance of LEA surpasses that of existing methods on multiple benchmark fMRI decoding and encoding datasets. By integrating fMRI decoding and encoding, LEA offers a comprehensive solution for modeling the intricate relationship between brain activity and visual stimuli.
comment: 12 pages, 5 figures
♻ ☆ InDL: A New Dataset and Benchmark for In-Diagram Logic Interpretation based on Visual Illusion
This paper introduces a novel approach to evaluating deep learning models' capacity for in-diagram logic interpretation. Leveraging the intriguing realm of visual illusions, we establish a unique dataset, InDL, designed to rigorously test and benchmark these models. Deep learning has witnessed remarkable progress in domains such as computer vision and natural language processing. However, models often stumble in tasks requiring logical reasoning due to their inherent 'black box' characteristics, which obscure the decision-making process. Our work presents a new lens to understand these models better by focusing on their handling of visual illusions -- a complex interplay of perception and logic. We utilize six classic geometric optical illusions to create a comparative framework between human and machine visual perception. This methodology offers a quantifiable measure to rank models, elucidating potential weaknesses and providing actionable insights for model improvements. Our experimental results affirm the efficacy of our benchmarking strategy, demonstrating its ability to effectively rank models based on their logic interpretation ability. As part of our commitment to reproducible research, the source code and datasets will be made publicly available at https://github.com/rabbit-magic-wh/InDL
comment: arXiv admin note: text overlap with arXiv:2305.02299, arXiv:2302.11939, arXiv:2301.13287, arXiv:2305.12686
♻ ☆ ImageCaptioner$^2$: Image Captioner for Image Captioning Bias Amplification Assessment
Most pre-trained learning systems are known to suffer from bias, which typically emerges from the data, the model, or both. Measuring and quantifying bias and its sources is a challenging task and has been extensively studied in image captioning. Despite the significant effort in this direction, we observed that existing metrics lack consistency in the inclusion of the visual signal. In this paper, we introduce a new bias assessment metric, dubbed $ImageCaptioner^2$, for image captioning. Instead of measuring the absolute bias in the model or the data, $ImageCaptioner^2$ pay more attention to the bias introduced by the model w.r.t the data bias, termed bias amplification. Unlike the existing methods, which only evaluate the image captioning algorithms based on the generated captions only, $ImageCaptioner^2$ incorporates the image while measuring the bias. In addition, we design a formulation for measuring the bias of generated captions as prompt-based image captioning instead of using language classifiers. Finally, we apply our $ImageCaptioner^2$ metric across 11 different image captioning architectures on three different datasets, i.e., MS-COCO caption dataset, Artemis V1, and Artemis V2, and on three different protected attributes, i.e., gender, race, and emotions. Consequently, we verify the effectiveness of our $ImageCaptioner^2$ metric by proposing AnonymousBench, which is a novel human evaluation paradigm for bias metrics. Our metric shows significant superiority over the recent bias metric; LIC, in terms of human alignment, where the correlation scores are 80% and 54% for our metric and LIC, respectively. The code is available at https://eslambakr.github.io/imagecaptioner2.github.io/.
♻ ☆ ICDAR 2023 Competition on Reading the Seal Title ICDAR2023
Reading seal title text is a challenging task due to the variable shapes of seals, curved text, background noise, and overlapped text. However, this important element is commonly found in official and financial scenarios, and has not received the attention it deserves in the field of OCR technology. To promote research in this area, we organized ICDAR 2023 competition on reading the seal title (ReST), which included two tasks: seal title text detection (Task 1) and end-to-end seal title recognition (Task 2). We constructed a dataset of 10,000 real seal data, covering the most common classes of seals, and labeled all seal title texts with text polygons and text contents. The competition opened on 30th December, 2022 and closed on 20th March, 2023. The competition attracted 53 participants from academia and industry including 28 submissions for Task 1 and 25 submissions for Task 2, which demonstrated significant interest in this challenging task. In this report, we present an overview of the competition, including the organization, challenges, and results. We describe the dataset and tasks, and summarize the submissions and evaluation results. The results show that significant progress has been made in the field of seal title text reading, and we hope that this competition will inspire further research and development in this important area of OCR technology.
comment: ICDAR2023 Competition on ReST report (To be appear in ICDAR 2023)
♻ ☆ Semantic Embedded Deep Neural Network: A Generic Approach to Boost Multi-Label Image Classification Performance
Fine-grained multi-label classification models have broad applications in e-commerce, such as visual based label predictions ranging from fashion attribute detection to brand recognition. One challenge to achieve satisfactory performance for those classification tasks in real world is the wild visual background signal that contains irrelevant pixels which confuses model to focus onto the region of interest and make prediction upon the specific region. In this paper, we introduce a generic semantic-embedding deep neural network to apply the spatial awareness semantic feature incorporating a channel-wise attention based model to leverage the localization guidance to boost model performance for multi-label prediction. We observed an Avg.relative improvement of 15.27% in terms of AUC score across all labels compared to the baseline approach. Core experiment and ablation studies involve multi-label fashion attribute classification performed on Instagram fashion apparels' image. We compared the model performances among our approach, baseline approach, and 3 alternative approaches to leverage semantic features. Results show favorable performance for our approach.
♻ ☆ Generated Faces in the Wild: Quantitative Comparison of Stable Diffusion, Midjourney and DALL-E 2
The field of image synthesis has made great strides in the last couple of years. Recent models are capable of generating images with astonishing quality. Fine-grained evaluation of these models on some interesting categories such as faces is still missing. Here, we conduct a quantitative comparison of three popular systems including Stable Diffusion, Midjourney, and DALL-E 2 in their ability to generate photorealistic faces in the wild. We find that Stable Diffusion generates better faces than the other systems, according to the FID score. We also introduce a dataset of generated faces in the wild dubbed GFW, including a total of 15,076 faces. Furthermore, we hope that our study spurs follow-up research in assessing the generative models and improving them. Data and code are available at data and code, respectively.
comment: dataset link udated!
♻ ☆ QuantArt: Quantizing Image Style Transfer Towards High Visual Fidelity CVPR 2023
The mechanism of existing style transfer algorithms is by minimizing a hybrid loss function to push the generated image toward high similarities in both content and style. However, this type of approach cannot guarantee visual fidelity, i.e., the generated artworks should be indistinguishable from real ones. In this paper, we devise a new style transfer framework called QuantArt for high visual-fidelity stylization. QuantArt pushes the latent representation of the generated artwork toward the centroids of the real artwork distribution with vector quantization. By fusing the quantized and continuous latent representations, QuantArt allows flexible control over the generated artworks in terms of content preservation, style similarity, and visual fidelity. Experiments on various style transfer settings show that our QuantArt framework achieves significantly higher visual fidelity compared with the existing style transfer methods.
comment: Accepted to CVPR 2023. Code is available at https://github.com/siyuhuang/QuantArt
Information Retrieval 19
☆ Learning Similarity among Users for Personalized Session-Based Recommendation from hierarchical structure of User-Session-Item
The task of the session-based recommendation is to predict the next interaction of the user based on the anonymized user's behavior pattern. And personalized version of this system is a promising research field due to its availability to deal with user information. However, there's a problem that the user's preferences and historical sessions were not considered in the typical session-based recommendation since it concentrates only on user-item interaction. In addition, the existing personalized session-based recommendation model has a limited capability in that it only considers the preference of the current user without considering those of similar users. It means there can be the loss of information included within the hierarchical data structure of the user-session-item. To tackle with this problem, we propose USP-SBR(abbr. of User Similarity Powered - Session Based Recommender). To model global historical sessions of users, we propose UserGraph that has two types of nodes - ItemNode and UserNode. We then connect the nodes with three types of edges. The first type of edges connects ItemNode as chronological order, and the second connects ItemNode to UserNode, and the last connects UserNode to ItemNode. With these user embeddings, we propose additional contrastive loss, that makes users with similar intention be close to each other in the vector space. we apply graph neural network on these UserGraph and update nodes. Experimental results on two real-world datasets demonstrate that our method outperforms some state-of-the-art approaches.
comment: 7 pages, 5 figures
☆ Gen-IR @ SIGIR 2023: The First Workshop on Generative Information Retrieval SIGIR 23
Generative information retrieval (IR) has experienced substantial growth across multiple research communities (e.g., information retrieval, computer vision, natural language processing, and machine learning), and has been highly visible in the popular press. Theoretical, empirical, and actual user-facing products have been released that retrieve documents (via generation) or directly generate answers given an input request. We would like to investigate whether end-to-end generative models are just another trend or, as some claim, a paradigm change for IR. This necessitates new metrics, theoretical grounding, evaluation methods, task definitions, models, user interfaces, etc. The goal of this workshop (https://coda.io/@sigir/gen-ir) is to focus on previously explored Generative IR techniques like document retrieval and direct Grounded Answer Generation, while also offering a venue for the discussion and exploration of how Generative IR can be applied to new domains like recommendation systems, summarization, etc. The format of the workshop is interactive, including roundtable and keynote sessions and tends to avoid the one-sided dialogue of a mini-conference.
comment: Accepted SIGIR 23 workshop
☆ Benchmarking Middle-Trained Language Models for Neural Search
Middle training methods aim to bridge the gap between the Masked Language Model (MLM) pre-training and the final finetuning for retrieval. Recent models such as CoCondenser, RetroMAE, and LexMAE argue that the MLM task is not sufficient enough to pre-train a transformer network for retrieval and hence propose various tasks to do so. Intrigued by those novel methods, we noticed that all these models used different finetuning protocols, making it hard to assess the benefits of middle training. We propose in this paper a benchmark of CoCondenser, RetroMAE, and LexMAE, under the same finetuning conditions. We compare both dense and sparse approaches under various finetuning protocols and middle training on different collections (MS MARCO, Wikipedia or Tripclick). We use additional middle training baselines, such as a standard MLM finetuning on the retrieval collection, optionally augmented by a CLS predicting the passage term frequency. For the sparse approach, our study reveals that there is almost no statistical difference between those methods: the more effective the finetuning procedure is, the less difference there is between those models. For the dense approach, RetroMAE using MS MARCO as middle-training collection shows excellent results in almost all the settings. Finally, we show that middle training on the retrieval collection, thus adapting the language model to it, is a critical factor. Overall, a better experimental setup should be adopted to evaluate middle training methods. Code available at https://github.com/naver/splade/tree/benchmarch-SIGIR23
☆ CTRL: Connect Tabular and Language Model for CTR Prediction
Traditional click-through rate (CTR) prediction models convert the tabular data into one-hot vectors and leverage the collaborative relations among features for inferring user's preference over items. This modeling paradigm discards the essential semantic information. Though some recent works like P5 and M6-Rec have explored the potential of using Pre-trained Language Models (PLMs) to extract semantic signals for CTR prediction, they are computationally expensive and suffer from low efficiency. Besides, the beneficial collaborative relations are not considered, hindering the recommendation performance. To solve these problems, in this paper, we propose a novel framework \textbf{CTRL}, which is industrial friendly and model-agnostic with high training and inference efficiency. Specifically, the original tabular data is first converted into textual data. Both tabular data and converted textual data are regarded as two different modalities and are separately fed into the collaborative CTR model and pre-trained language model. A cross-modal knowledge alignment procedure is performed to fine-grained align and integrate the collaborative and semantic signals, and the lightweight collaborative model can be deployed online for efficient serving after fine-tuned with supervised signals. Experimental results on three public datasets show that CTRL outperforms the SOTA CTR models significantly. Moreover, we further verify its effectiveness on a large-scale industrial recommender system.
☆ Path-Specific Counterfactual Fairness for Recommender Systems KDD 2023
Recommender systems (RSs) have become an indispensable part of online platforms. With the growing concerns of algorithmic fairness, RSs are not only expected to deliver high-quality personalized content, but are also demanded not to discriminate against users based on their demographic information. However, existing RSs could capture undesirable correlations between sensitive features and observed user behaviors, leading to biased recommendations. Most fair RSs tackle this problem by completely blocking the influences of sensitive features on recommendations. But since sensitive features may also affect user interests in a fair manner (e.g., race on culture-based preferences), indiscriminately eliminating all the influences of sensitive features inevitably degenerate the recommendations quality and necessary diversities. To address this challenge, we propose a path-specific fair RS (PSF-RS) for recommendations. Specifically, we summarize all fair and unfair correlations between sensitive features and observed ratings into two latent proxy mediators, where the concept of path-specific bias (PS-Bias) is defined based on path-specific counterfactual inference. Inspired by Pearl's minimal change principle, we address the PS-Bias by minimally transforming the biased factual world into a hypothetically fair world, where a fair RS model can be learned accordingly by solving a constrained optimization problem. For the technical part, we propose a feasible implementation of PSF-RS, i.e., PSF-VAE, with weakly-supervised variational inference, which robustly infers the latent mediators such that unfairness can be mitigated while necessary recommendation diversities can be maximally preserved simultaneously. Experiments conducted on semi-simulated and real-world datasets demonstrate the effectiveness of PSF-RS.
comment: Accepted by KDD 2023 research track
☆ Learning to Relate to Previous Turns in Conversational Search KDD 2023
Conversational search allows a user to interact with a search system in multiple turns. A query is strongly dependent on the conversation context. An effective way to improve retrieval effectiveness is to expand the current query with historical queries. However, not all the previous queries are related to, and useful for expanding the current query. In this paper, we propose a new method to select relevant historical queries that are useful for the current query. To cope with the lack of labeled training data, we use a pseudo-labeling approach to annotate useful historical queries based on their impact on the retrieval results. The pseudo-labeled data are used to train a selection model. We further propose a multi-task learning framework to jointly train the selector and the retriever during fine-tuning, allowing us to mitigate the possible inconsistency between the pseudo labels and the changed retriever. Extensive experiments on four conversational search datasets demonstrate the effectiveness and broad applicability of our method compared with several strong baselines.
comment: Accepted by SIGKDD 2023 Research Track
☆ RecAgent: A Novel Simulation Paradigm for Recommender Systems
Recommender system has deeply revolutionized people's daily life and production, bringing a large amount of business value. In the recommendation domain, simulation and real data-based studies are two typical research paradigms, with each having different advantages. Previously, real data-based studies occupy more important positions, since accurately simulating the user preference is quite difficult. Recently, large language models (LLM) have shown great potential to achieve human-like intelligence, which provides new opportunities to overcome the shortcomings of simulation-based studies and thus highlight their advantages, such as much more application scenarios and cheaper data acquisition strategies. To shed lights on this direction, in this paper, we introduce an LLM-based recommender simulator called RecAgent. Our simulator is composed of two modules: (1) the user module and (2) the recommender module. The user module can browse the recommendation website, communicate with other users and broadcast messages on the social media. The recommender module is designed to provide search or recommendation lists to the users, and one can design different models to implement the recommender. All the users take actions based on LLMs, and can freely evolve like in the real world. We present several case studies to demonstrate that the users in our simulator can indeed behave in a reasonable manner as expected. Our project has been released at https://github.com/RUC-GSAI/YuLan-Rec.
comment: 12 pages, 2 figures
☆ Evaluation of AI Chatbots for Patient-Specific EHR Questions
This paper investigates the use of artificial intelligence chatbots for patient-specific question answering (QA) from clinical notes using several large language model (LLM) based systems: ChatGPT (versions 3.5 and 4), Google Bard, and Claude. We evaluate the accuracy, relevance, comprehensiveness, and coherence of the answers generated by each model using a 5-point Likert scale on a set of patient-specific questions.
☆ SamToNe: Improving Contrastive Loss for Dual Encoder Retrieval Models with Same Tower Negatives ACL 2023
Dual encoders have been used for retrieval tasks and representation learning with good results. A standard way to train dual encoders is using a contrastive loss with in-batch negatives. In this work, we propose an improved contrastive learning objective by adding queries or documents from the same encoder towers to the negatives, for which we name it as "contrastive loss with SAMe TOwer NEgatives" (SamToNe). By evaluating on question answering retrieval benchmarks from MS MARCO and MultiReQA, and heterogenous zero-shot information retrieval benchmarks (BEIR), we demonstrate that SamToNe can effectively improve the retrieval quality for both symmetric and asymmetric dual encoders. By directly probing the embedding spaces of the two encoding towers via the t-SNE algorithm (van der Maaten and Hinton, 2008), we observe that SamToNe ensures the alignment between the embedding spaces from the two encoder towers. Based on the analysis of the embedding distance distributions of the top-$1$ retrieved results, we further explain the efficacy of the method from the perspective of regularisation.
comment: ACL 2023 Findings
☆ Construction d'un système de recommandation basé sur des contraintes via des graphes de connaissances
Knowledge graphs in RDF model entities and their relations using ontologies, and have gained popularity for information modeling. In recommender systems, knowledge graphs help represent more links and relationships between users and items. Constraint-based recommender systems leverage deep recommendation knowledge to identify relevant suggestions. When combined with knowledge graphs, they offer benefits in constraint sets. This paper explores a constraint-based recommender system using RDF knowledge graphs for the vehicle purchase/sale domain. Our experiments demonstrate that the proposed approach efficiently identifies recommendations based on user preferences.
comment: in French language
☆ Lumos in the Night Sky: AI-enabled Visual Tool for Exploring Night-Time Light Patterns ECML
We introduce NightPulse, an interactive tool for Night-time light (NTL) data visualization and analytics, which enables researchers and stakeholders to explore and analyze NTL data with a user-friendly platform. Powered by efficient system architecture, NightPulse supports image segmentation, clustering, and change pattern detection to identify urban development and sprawl patterns. It captures temporal trends of NTL and semantics of cities, answering questions about demographic factors, city boundaries, and unusual differences.
comment: 5 pages, 3 figures. Accepted in ECML PKDD Demo track
☆ Personalized Federated Domain Adaptation for Item-to-Item Recommendation UAI 2023
Item-to-Item (I2I) recommendation is an important function in most recommendation systems, which generates replacement or complement suggestions for a particular item based on its semantic similarities to other cataloged items. Given that subsets of items in a recommendation system might be co-interacted with by the same set of customers, graph-based models, such as graph neural networks (GNNs), provide a natural framework to combine, ingest and extract valuable insights from such high-order relational interactions between cataloged items, as well as their metadata features, as has been shown in many recent studies. However, learning GNNs effectively for I2I requires ingesting a large amount of relational data, which might not always be available, especially in new, emerging market segments. To mitigate this data bottleneck, we postulate that recommendation patterns learned from existing mature market segments (with private data) could be adapted to build effective warm-start models for emerging ones. To achieve this, we propose and investigate a personalized federated modeling framework based on GNNs to summarize, assemble and adapt recommendation patterns across market segments with heterogeneous customer behaviors into effective local models. Our key contribution is a personalized graph adaptation model that bridges the gap between recent literature on federated GNNs and (non-graph) personalized federated learning, which either does not optimize for the adaptability of the federated model or is restricted to local models with homogeneous parameterization, excluding GNNs with heterogeneous local graphs.
comment: 14 pages, Accepted by UAI 2023
☆ Unsupervised Dense Retrieval with Relevance-Aware Contrastive Pre-Training ACL 2023
Dense retrievers have achieved impressive performance, but their demand for abundant training data limits their application scenarios. Contrastive pre-training, which constructs pseudo-positive examples from unlabeled data, has shown great potential to solve this problem. However, the pseudo-positive examples crafted by data augmentations can be irrelevant. To this end, we propose relevance-aware contrastive learning. It takes the intermediate-trained model itself as an imperfect oracle to estimate the relevance of positive pairs and adaptively weighs the contrastive loss of different pairs according to the estimated relevance. Our method consistently improves the SOTA unsupervised Contriever model on the BEIR and open-domain QA retrieval benchmarks. Further exploration shows that our method can not only beat BM25 after further pre-training on the target corpus but also serves as a good few-shot learner. Our code is publicly available at https://github.com/Yibin-Lei/ReContriever.
comment: ACL 2023 Findings (Short), 5 pages main + 1 page references + 1 page appendix
♻ ☆ Tensor Completion with Provable Consistency and Fairness Guarantees for Recommender Systems
We introduce a new consistency-based approach for defining and solving nonnegative/positive matrix and tensor completion problems. The novelty of the framework is that instead of artificially making the problem well-posed in the form of an application-arbitrary optimization problem, e.g., minimizing a bulk structural measure such as rank or norm, we show that a single property/constraint: preserving unit-scale consistency, guarantees the existence of both a solution and, under relatively weak support assumptions, uniqueness. The framework and solution algorithms also generalize directly to tensors of arbitrary dimensions while maintaining computational complexity that is linear in problem size for fixed dimension d. In the context of recommender system (RS) applications, we prove that two reasonable properties that should be expected to hold for any solution to the RS problem are sufficient to permit uniqueness guarantees to be established within our framework. Key theoretical contributions include a general unit-consistent tensor-completion framework with proofs of its properties, e.g., consensus-order and fairness, and algorithms with optimal runtime and space complexities, e.g., O(1) term-completion with preprocessing complexity that is linear in the number of known terms of the matrix/tensor. From a practical perspective, the seamless ability of the framework to generalize to exploit high-dimensional structural relationships among key state variables, e.g., user and product attributes, offers a means for extracting significantly more information than is possible for alternative methods that cannot generalize beyond direct user-product relationships. Finally, we propose our consensus ordering property as an admissibility criterion for any proposed RS method.
comment: Final revision after acceptance by journal
♻ ☆ Candidate Set Re-ranking for Composed Image Retrieval with Dual Multi-modal Encoder
Composed image retrieval aims to find an image that best matches a given multi-modal user query consisting of a reference image and text pair. Existing methods commonly pre-compute image embeddings over the entire corpus and compare these to a reference image embedding modified by the query text at test time. Such a pipeline is very efficient at test time since fast vector distances can be used to evaluate candidates, but modifying the reference image embedding guided only by a short textual description can be difficult, especially independent of potential candidates. An alternative approach is to allow interactions between the query and every possible candidate, i.e., reference-text-candidate triplets, and pick the best from the entire set. Though this approach is more discriminative, for large-scale datasets the computational cost is prohibitive since pre-computation of candidate embeddings is no longer possible. We propose to combine the merits of both schemes using a two-stage model. Our first stage adopts the conventional vector distancing metric and performs a fast pruning among candidates. Meanwhile, our second stage employs a dual-encoder architecture, which effectively attends to the input triplet of reference-text-candidate and re-ranks the candidates. Both stages utilize a vision-and-language pre-trained network, which has proven beneficial for various downstream tasks. Our method consistently outperforms state-of-the-art approaches on standard benchmarks for the task.
comment: 14 pages
♻ ☆ AdaTT: Adaptive Task-to-Task Fusion Network for Multitask Learning in Recommendations
Multi-task learning (MTL) aims to enhance the performance and efficiency of machine learning models by simultaneously training them on multiple tasks. However, MTL research faces two challenges: 1) effectively modeling the relationships between tasks to enable knowledge sharing, and 2) jointly learning task-specific and shared knowledge. In this paper, we present a novel model called Adaptive Task-to-Task Fusion Network (AdaTT) to address both challenges. AdaTT is a deep fusion network built with task-specific and optional shared fusion units at multiple levels. By leveraging a residual mechanism and a gating mechanism for task-to-task fusion, these units adaptively learn both shared knowledge and task-specific knowledge. To evaluate AdaTT's performance, we conduct experiments on a public benchmark and an industrial recommendation dataset using various task groups. Results demonstrate AdaTT significantly outperforms existing state-of-the-art baselines. Furthermore, our end-to-end experiments reveal that the model exhibits better performance compared to alternatives.
♻ ☆ Multi-hop Evidence Retrieval for Cross-document Relation Extraction ACL 2023
Relation Extraction (RE) has been extended to cross-document scenarios because many relations are not simply described in a single document. This inevitably brings the challenge of efficient open-space evidence retrieval to support the inference of cross-document relations, along with the challenge of multi-hop reasoning on top of entities and evidence scattered in an open set of documents. To combat these challenges, we propose MR.COD (Multi-hop evidence retrieval for Cross-document relation extraction), which is a multi-hop evidence retrieval method based on evidence path mining and ranking. We explore multiple variants of retrievers to show evidence retrieval is essential in cross-document RE. We also propose a contextual dense retriever for this setting. Experiments on CodRED show that evidence retrieval with MR.COD effectively acquires crossdocument evidence and boosts end-to-end RE performance in both closed and open settings.
comment: ACL 2023 (Findings)
♻ ☆ Empowering Long-tail Item Recommendation through Cross Decoupling Network (CDN) KDD 2023
Industry recommender systems usually suffer from highly-skewed long-tail item distributions where a small fraction of the items receives most of the user feedback. This skew hurts recommender quality especially for the item slices without much user feedback. While there have been many research advances made in academia, deploying these methods in production is very difficult and very few improvements have been made in industry. One challenge is that these methods often hurt overall performance; additionally, they could be complex and expensive to train and serve. In this work, we aim to improve tail item recommendations while maintaining the overall performance with less training and serving cost. We first find that the predictions of user preferences are biased under long-tail distributions. The bias comes from the differences between training and serving data in two perspectives: 1) the item distributions, and 2) user's preference given an item. Most existing methods mainly attempt to reduce the bias from the item distribution perspective, ignoring the discrepancy from user preference given an item. This leads to a severe forgetting issue and results in sub-optimal performance. To address the problem, we design a novel Cross Decoupling Network (CDN) (i) decouples the learning process of memorization and generalization on the item side through a mixture-of-expert architecture; (ii) decouples the user samples from different distributions through a regularized bilateral branch network. Finally, a new adapter is introduced to aggregate the decoupled vectors, and softly shift the training attention to tail items. Extensive experimental results show that CDN significantly outperforms state-of-the-art approaches on benchmark datasets. We also demonstrate its effectiveness by a case study of CDN in a large-scale recommendation system at Google.
comment: Accepted by KDD 2023 Applied Data Science (ADS) track
♻ ☆ Learning Personalized Page Content Ranking Using Customer Representation
On E-commerce stores, there are rich recommendation content to help shoppers shopping more efficiently. However given numerous products, it's crucial to select most relevant content to reduce the burden of information overload. We introduced a content ranking service powered by a linear causal bandit algorithm to rank and select content for each shopper under each context. The algorithm mainly leverages aggregated customer behavior features, and ignores single shopper level past activities. We study the problem of inferring shoppers interest from historical activities. We propose a deep learning based bandit algorithm that incorporates historical shopping behavior, customer latent shopping goals, and the correlation between customers and content categories. This model produces more personalized content ranking measured by 12.08% nDCG lift.
Machine Learning 150
☆ DeepGraphDMD: Interpretable Spatio-Temporal Decomposition of Non-linear Functional Brain Network Dynamics MICCAI 2023
Functional brain dynamics is supported by parallel and overlapping functional network modes that are associated with specific neural circuits. Decomposing these network modes from fMRI data and finding their temporal characteristics is challenging due to their time-varying nature and the non-linearity of the functional dynamics. Dynamic Mode Decomposition (DMD) algorithms have been quite popular for solving this decomposition problem in recent years. In this work, we apply GraphDMD -- an extension of the DMD for network data -- to extract the dynamic network modes and their temporal characteristics from the fMRI time series in an interpretable manner. GraphDMD, however, regards the underlying system as a linear dynamical system that is sub-optimal for extracting the network modes from non-linear functional data. In this work, we develop a generalized version of the GraphDMD algorithm -- DeepGraphDMD -- applicable to arbitrary non-linear graph dynamical systems. DeepGraphDMD is an autoencoder-based deep learning model that learns Koopman eigenfunctions for graph data and embeds the non-linear graph dynamics into a latent linear space. We show the effectiveness of our method in both simulated data and the HCP resting-state fMRI data. In the HCP data, DeepGraphDMD provides novel insights into cognitive brain functions by discovering two major network modes related to fluid and crystallized intelligence.
comment: to be published in MICCAI 2023
☆ SpQR: A Sparse-Quantized Representation for Near-Lossless LLM Weight Compression
Recent advances in large language model (LLM) pretraining have led to high-quality LLMs with impressive abilities. By compressing such LLMs via quantization to 3-4 bits per parameter, they can fit into memory-limited devices such as laptops and mobile phones, enabling personalized use. However, quantization down to 3-4 bits per parameter usually leads to moderate-to-high accuracy losses, especially for smaller models in the 1-10B parameter range, which are well-suited for edge deployments. To address this accuracy issue, we introduce the Sparse-Quantized Representation (SpQR), a new compressed format and quantization technique which enables for the first time near-lossless compression of LLMs across model scales, while reaching similar compression levels to previous methods. SpQR works by identifying and isolating outlier weights, which cause particularly-large quantization errors, and storing them in higher precision, while compressing all other weights to 3-4 bits, and achieves relative accuracy losses of less than 1% in perplexity for highly-accurate LLaMA and Falcon LLMs. This makes it possible to run 33B parameter LLM on a single 24 GB consumer GPU without any performance degradation at 15% speedup thus making powerful LLMs available to consumer without any downsides. SpQR comes with efficient algorithms for both encoding weights into its format, as well as decoding them efficiently at runtime. Specifically, we provide an efficient GPU inference algorithm for SpQR which yields faster inference than 16-bit baselines at similar accuracy, while enabling memory compression gains of more than 4x.
comment: Extended preprint
☆ Sensitivity-Aware Finetuning for Accuracy Recovery on Deep Learning Hardware
Existing methods to recover model accuracy on analog-digital hardware in the presence of quantization and analog noise include noise-injection training. However, it can be slow in practice, incurring high computational costs, even when starting from pretrained models. We introduce the Sensitivity-Aware Finetuning (SAFT) approach that identifies noise sensitive layers in a model, and uses the information to freeze specific layers for noise-injection training. Our results show that SAFT achieves comparable accuracy to noise-injection training and is 2x to 8x faster.
comment: 7 pages, 2 figures
☆ A General Perspective on Objectives of Reinforcement Learning
In this lecture, we present a general perspective on reinforcement learning (RL) objectives, where we show three versions of objectives. The first version is the standard definition of objective in RL literature. Then we extend the standard definition to the $\lambda$-return version, which unifies the standard definition of objective. Finally, we propose a general objective that unifies the previous two versions. The last version provides a high level to understand of RL's objective, where it shows a fundamental formulation that connects some widely used RL techniques (e.g., TD$(\lambda)$ and GAE), and this objective can be potentially applied to extensive RL algorithms.
☆ Explore to Generalize in Zero-Shot RL
We study zero-shot generalization in reinforcement learning - optimizing a policy on a set of training tasks such that it will perform well on a similar but unseen test task. To mitigate overfitting, previous work explored different notions of invariance to the task. However, on problems such as the ProcGen Maze, an adequate solution that is invariant to the task visualization does not exist, and therefore invariance-based approaches fail. Our insight is that learning a policy that $\textit{explores}$ the domain effectively is harder to memorize than a policy that maximizes reward for a specific task, and therefore we expect such learned behavior to generalize well; we indeed demonstrate this empirically on several domains that are difficult for invariance-based approaches. Our $\textit{Explore to Generalize}$ algorithm (ExpGen) builds on this insight: We train an additional ensemble of agents that optimize reward. At test time, either the ensemble agrees on an action, and we generalize well, or we take exploratory actions, which are guaranteed to generalize and drive us to a novel part of the state space, where the ensemble may potentially agree again. We show that our approach is the state-of-the-art on several tasks in the ProcGen challenge that have so far eluded effective generalization. For example, we demonstrate a success rate of $82\%$ on the Maze task and $74\%$ on Heist with $200$ training levels.
☆ Of Mice and Mates: Automated Classification and Modelling of Mouse Behaviour in Groups using a Single Model across Cages
Behavioural experiments often happen in specialised arenas, but this may confound the analysis. To address this issue, we provide tools to study mice in the homecage environment, equipping biologists with the possibility to capture the temporal aspect of the individual's behaviour and model the interaction and interdependence between cage-mates with minimal human intervention. We develop the Activity Labelling Module (ALM) to automatically classify mouse behaviour from video, and a novel Group Behaviour Model (GBM) for summarising their joint behaviour across cages, using a permutation matrix to match the mouse identities in each cage to the model. We also release two datasets, ABODe for training behaviour classifiers and IMADGE for modelling behaviour.
☆ LibAUC: A Deep Learning Library for X-Risk Optimization KDD2023
This paper introduces the award-winning deep learning (DL) library called LibAUC for implementing state-of-the-art algorithms towards optimizing a family of risk functions named X-risks. X-risks refer to a family of compositional functions in which the loss function of each data point is defined in a way that contrasts the data point with a large number of others. They have broad applications in AI for solving classical and emerging problems, including but not limited to classification for imbalanced data (CID), learning to rank (LTR), and contrastive learning of representations (CLR). The motivation of developing LibAUC is to address the convergence issues of existing libraries for solving these problems. In particular, existing libraries may not converge or require very large mini-batch sizes in order to attain good performance for these problems, due to the usage of the standard mini-batch technique in the empirical risk minimization (ERM) framework. Our library is for deep X-risk optimization (DXO) that has achieved great success in solving a variety of tasks for CID, LTR and CLR. The contributions of this paper include: (1) It introduces a new mini-batch based pipeline for implementing DXO algorithms, which differs from existing DL pipeline in the design of controlled data samplers and dynamic mini-batch losses; (2) It provides extensive benchmarking experiments for ablation studies and comparison with existing libraries. The LibAUC library features scalable performance for millions of items to be contrasted, faster and better convergence than existing libraries for optimizing X-risks, seamless PyTorch deployment and versatile APIs for various loss optimization. Our library is available to the open source community at https://github.com/Optimization-AI/LibAUC, to facilitate further academic research and industrial applications.
comment: Accepted by KDD2023
☆ Discriminative Adversarial Privacy: Balancing Accuracy and Membership Privacy in Neural Networks
The remarkable proliferation of deep learning across various industries has underscored the importance of data privacy and security in AI pipelines. As the evolution of sophisticated Membership Inference Attacks (MIAs) threatens the secrecy of individual-specific information used for training deep learning models, Differential Privacy (DP) raises as one of the most utilized techniques to protect models against malicious attacks. However, despite its proven theoretical properties, DP can significantly hamper model performance and increase training time, turning its use impractical in real-world scenarios. Tackling this issue, we present Discriminative Adversarial Privacy (DAP), a novel learning technique designed to address the limitations of DP by achieving a balance between model performance, speed, and privacy. DAP relies on adversarial training based on a novel loss function able to minimise the prediction error while maximising the MIA's error. In addition, we introduce a novel metric named Accuracy Over Privacy (AOP) to capture the performance-privacy trade-off. Finally, to validate our claims, we compare DAP with diverse DP scenarios, providing an analysis of the results from performance, time, and privacy preservation perspectives.
☆ Forecasting Crude Oil Prices Using Reservoir Computing Models
Accurate crude oil price prediction is crucial for financial decision-making. We propose a novel reservoir computing model for forecasting crude oil prices. It outperforms popular deep learning methods in most scenarios, as demonstrated through rigorous evaluation using daily closing price data from major stock market indices. Our model's competitive advantage is further validated by comparing it with recent deep-learning approaches. This study introduces innovative reservoir computing models for predicting crude oil prices, with practical implications for financial practitioners. By leveraging advanced techniques, market participants can enhance decision-making and gain valuable insights into crude oil market dynamics.
comment: 14 pages, 4 figures
☆ SERT: A Transfomer Based Model for Spatio-Temporal Sensor Data with Missing Values for Environmental Monitoring
Environmental monitoring is crucial to our understanding of climate change, biodiversity loss and pollution. The availability of large-scale spatio-temporal data from sources such as sensors and satellites allows us to develop sophisticated models for forecasting and understanding key drivers. However, the data collected from sensors often contain missing values due to faulty equipment or maintenance issues. The missing values rarely occur simultaneously leading to data that are multivariate misaligned sparse time series. We propose two models that are capable of performing multivariate spatio-temporal forecasting while handling missing data naturally without the need for imputation. The first model is a transformer-based model, which we name SERT (Spatio-temporal Encoder Representations from Transformers). The second is a simpler model named SST-ANN (Sparse Spatio-Temporal Artificial Neural Network) which is capable of providing interpretable results. We conduct extensive experiments on two different datasets for multivariate spatio-temporal forecasting and show that our models have competitive or superior performance to those at the state-of-the-art.
comment: 11 pages, 7 figures
☆ Learning Similarity among Users for Personalized Session-Based Recommendation from hierarchical structure of User-Session-Item
The task of the session-based recommendation is to predict the next interaction of the user based on the anonymized user's behavior pattern. And personalized version of this system is a promising research field due to its availability to deal with user information. However, there's a problem that the user's preferences and historical sessions were not considered in the typical session-based recommendation since it concentrates only on user-item interaction. In addition, the existing personalized session-based recommendation model has a limited capability in that it only considers the preference of the current user without considering those of similar users. It means there can be the loss of information included within the hierarchical data structure of the user-session-item. To tackle with this problem, we propose USP-SBR(abbr. of User Similarity Powered - Session Based Recommender). To model global historical sessions of users, we propose UserGraph that has two types of nodes - ItemNode and UserNode. We then connect the nodes with three types of edges. The first type of edges connects ItemNode as chronological order, and the second connects ItemNode to UserNode, and the last connects UserNode to ItemNode. With these user embeddings, we propose additional contrastive loss, that makes users with similar intention be close to each other in the vector space. we apply graph neural network on these UserGraph and update nodes. Experimental results on two real-world datasets demonstrate that our method outperforms some state-of-the-art approaches.
comment: 7 pages, 5 figures
☆ AI Techniques for Cone Beam Computed Tomography in Dentistry: Trends and Practices
Cone-beam computed tomography (CBCT) is a popular imaging modality in dentistry for diagnosing and planning treatment for a variety of oral diseases with the ability to produce detailed, three-dimensional images of the teeth, jawbones, and surrounding structures. CBCT imaging has emerged as an essential diagnostic tool in dentistry. CBCT imaging has seen significant improvements in terms of its diagnostic value, as well as its accuracy and efficiency, with the most recent development of artificial intelligence (AI) techniques. This paper reviews recent AI trends and practices in dental CBCT imaging. AI has been used for lesion detection, malocclusion classification, measurement of buccal bone thickness, and classification and segmentation of teeth, alveolar bones, mandibles, landmarks, contours, and pharyngeal airways using CBCT images. Mainly machine learning algorithms, deep learning algorithms, and super-resolution techniques are used for these tasks. This review focuses on the potential of AI techniques to transform CBCT imaging in dentistry, which would improve both diagnosis and treatment planning. Finally, we discuss the challenges and limitations of artificial intelligence in dentistry and CBCT imaging.
comment: Recent Advances in Electrical, Electronics & Digital Healthcare Technologies REEDCON 2023
☆ Interpretable Alzheimer's Disease Classification Via a Contrastive Diffusion Autoencoder
In visual object classification, humans often justify their choices by comparing objects to prototypical examples within that class. We may therefore increase the interpretability of deep learning models by imbuing them with a similar style of reasoning. In this work, we apply this principle by classifying Alzheimer's Disease based on the similarity of images to training examples within the latent space. We use a contrastive loss combined with a diffusion autoencoder backbone, to produce a semantically meaningful latent space, such that neighbouring latents have similar image-level features. We achieve a classification accuracy comparable to black box approaches on a dataset of 2D MRI images, whilst producing human interpretable model explanations. Therefore, this work stands as a contribution to the pertinent development of accurate and interpretable deep learning within medical imaging.
☆ Automating Style Analysis and Visualization With Explainable AI -- Case Studies on Brand Recognition
Incorporating style-related objectives into shape design has been centrally important to maximize product appeal. However, stylistic features such as aesthetics and semantic attributes are hard to codify even for experts. As such, algorithmic style capture and reuse have not fully benefited from automated data-driven methodologies due to the challenging nature of design describability. This paper proposes an AI-driven method to fully automate the discovery of brand-related features. Our approach introduces BIGNet, a two-tier Brand Identification Graph Neural Network (GNN) to classify and analyze scalar vector graphics (SVG). First, to tackle the scarcity of vectorized product images, this research proposes two data acquisition workflows: parametric modeling from small curve-based datasets, and vectorization from large pixel-based datasets. Secondly, this study constructs a novel hierarchical GNN architecture to learn from both SVG's curve-level and chunk-level parameters. In the first case study, BIGNet not only classifies phone brands but also captures brand-related features across multiple scales, such as the location of the lens, the height-width ratio, and the screen-frame gap, as confirmed by AI evaluation. In the second study, this paper showcases the generalizability of BIGNet learning from a vectorized car image dataset and validates the consistency and robustness of its predictions given four scenarios. The results match the difference commonly observed in luxury vs. economy brands in the automobile market. Finally, this paper also visualizes the activation maps generated from a convolutional neural network and shows BIGNet's advantage of being a more human-friendly, explainable, and explicit style-capturing agent. Code and dataset can be found on Github: 1. Phone case study: github.com/parksandrecfan/bignet-phone 2. Car case study: github.com/parksandrecfan/bignet-car
☆ Quantification of Uncertainties in Deep Learning-based Environment Perception
In this work, we introduce a novel Deep Learning-based method to perceive the environment of a vehicle based on radar scans while accounting for uncertainties in its predictions. The environment of the host vehicle is segmented into equally sized grid cells which are classified individually. Complementary to the segmentation output, our Deep Learning-based algorithm is capable of differentiating uncertainties in its predictions as being related to an inadequate model (epistemic uncertainty) or noisy data (aleatoric uncertainty). To this end, weights are described as probability distributions accounting for uncertainties in the model parameters. Distributions are learned in a supervised fashion using gradient descent. We prove that uncertainties in the model output correlate with the precision of its predictions. Compared to previous concepts, we show superior performance of our approach to reliably perceive the environment of a vehicle.
comment: 2021 IEEE International Conference on Omni-Layer Intelligent Systems (COINS), Barcelona, Spain, 2021
☆ On the Behavior of Intrusive and Non-intrusive Speech Enhancement Metrics in Predictive and Generative Settings
Since its inception, the field of deep speech enhancement has been dominated by predictive (discriminative) approaches, such as spectral mapping or masking. Recently, however, novel generative approaches have been applied to speech enhancement, attaining good denoising performance with high subjective quality scores. At the same time, advances in deep learning also allowed for the creation of neural network-based metrics, which have desirable traits such as being able to work without a reference (non-intrusively). Since generatively enhanced speech tends to exhibit radically different residual distortions, its evaluation using instrumental speech metrics may behave differently compared to predictively enhanced speech. In this paper, we evaluate the performance of the same speech enhancement backbone trained under predictive and generative paradigms on a variety of metrics and show that intrusive and non-intrusive measures correlate differently for each paradigm. This analysis motivates the search for metrics that can together paint a complete and unbiased picture of speech enhancement performance, irrespective of the model's training process.
comment: Submitted to ITG Conference on Speech Communication
☆ Hiding in Plain Sight: Disguising Data Stealing Attacks in Federated Learning
Malicious server (MS) attacks have enabled the scaling of data stealing in federated learning to large batch sizes and secure aggregation, settings previously considered private. However, many concerns regarding client-side detectability of MS attacks were raised, questioning their practicality once they are publicly known. In this work, for the first time, we thoroughly study the problem of client-side detectability.We demonstrate that most prior MS attacks, which fundamentally rely on one of two key principles, are detectable by principled client-side checks. Further, we formulate desiderata for practical MS attacks and propose SEER, a novel attack framework that satisfies all desiderata, while stealing user data from gradients of realistic networks, even for large batch sizes (up to 512 in our experiments) and under secure aggregation. The key insight of SEER is the use of a secret decoder, which is jointly trained with the shared model. Our work represents a promising first step towards more principled treatment of MS attacks, paving the way for realistic data stealing that can compromise user privacy in real-world deployments.
☆ Interval Load Forecasting for Individual Households in the Presence of Electric Vehicle Charging
The transition to Electric Vehicles (EV) in place of traditional internal combustion engines is increasing societal demand for electricity. The ability to integrate the additional demand from EV charging into forecasting electricity demand is critical for maintaining the reliability of electricity generation and distribution. Load forecasting studies typically exclude households with home EV charging, focusing on offices, schools, and public charging stations. Moreover, they provide point forecasts which do not offer information about prediction uncertainty. Consequently, this paper proposes the Long Short-Term Memory Bayesian Neural Networks (LSTM-BNNs) for household load forecasting in presence of EV charging. The approach takes advantage of the LSTM model to capture the time dependencies and uses the dropout layer with Bayesian inference to generate prediction intervals. Results show that the proposed LSTM-BNNs achieve accuracy similar to point forecasts with the advantage of prediction intervals. Moreover, the impact of lockdowns related to the COVID-19 pandemic on the load forecasting model is examined, and the analysis shows that there is no major change in the model performance as, for the considered households, the randomness of the EV charging outweighs the change due to pandemic.
☆ Using Sequences of Life-events to Predict Human Lives
Over the past decade, machine learning has revolutionized computers' ability to analyze text through flexible computational models. Due to their structural similarity to written language, transformer-based architectures have also shown promise as tools to make sense of a range of multi-variate sequences from protein-structures, music, electronic health records to weather-forecasts. We can also represent human lives in a way that shares this structural similarity to language. From one perspective, lives are simply sequences of events: People are born, visit the pediatrician, start school, move to a new location, get married, and so on. Here, we exploit this similarity to adapt innovations from natural language processing to examine the evolution and predictability of human lives based on detailed event sequences. We do this by drawing on arguably the most comprehensive registry data in existence, available for an entire nation of more than six million individuals across decades. Our data include information about life-events related to health, education, occupation, income, address, and working hours, recorded with day-to-day resolution. We create embeddings of life-events in a single vector space showing that this embedding space is robust and highly structured. Our models allow us to predict diverse outcomes ranging from early mortality to personality nuances, outperforming state-of-the-art models by a wide margin. Using methods for interpreting deep learning models, we probe the algorithm to understand the factors that enable our predictions. Our framework allows researchers to identify new potential mechanisms that impact life outcomes and associated possibilities for personalized interventions.
☆ Nonparametric Iterative Machine Teaching ICML 2023
In this paper, we consider the problem of Iterative Machine Teaching (IMT), where the teacher provides examples to the learner iteratively such that the learner can achieve fast convergence to a target model. However, existing IMT algorithms are solely based on parameterized families of target models. They mainly focus on convergence in the parameter space, resulting in difficulty when the target models are defined to be functions without dependency on parameters. To address such a limitation, we study a more general task -- Nonparametric Iterative Machine Teaching (NIMT), which aims to teach nonparametric target models to learners in an iterative fashion. Unlike parametric IMT that merely operates in the parameter space, we cast NIMT as a functional optimization problem in the function space. To solve it, we propose both random and greedy functional teaching algorithms. We obtain the iterative teaching dimension (ITD) of the random teaching algorithm under proper assumptions, which serves as a uniform upper bound of ITD in NIMT. Further, the greedy teaching algorithm has a significantly lower ITD, which reaches a tighter upper bound of ITD in NIMT. Finally, we verify the correctness of our theoretical findings with extensive experiments in nonparametric scenarios.
comment: ICML 2023 (20 pages, 10 figures)
☆ Unveiling the Two-Faced Truth: Disentangling Morphed Identities for Face Morphing Detection
Morphing attacks keep threatening biometric systems, especially face recognition systems. Over time they have become simpler to perform and more realistic, as such, the usage of deep learning systems to detect these attacks has grown. At the same time, there is a constant concern regarding the lack of interpretability of deep learning models. Balancing performance and interpretability has been a difficult task for scientists. However, by leveraging domain information and proving some constraints, we have been able to develop IDistill, an interpretable method with state-of-the-art performance that provides information on both the identity separation on morph samples and their contribution to the final prediction. The domain information is learnt by an autoencoder and distilled to a classifier system in order to teach it to separate identity information. When compared to other methods in the literature it outperforms them in three out of five databases and is competitive in the remaining.
comment: Accepted at EUSIPCO 2023
☆ Over-the-Air Federated Learning in Satellite systems
Federated learning in satellites offers several advantages. Firstly, it ensures data privacy and security, as sensitive data remains on the satellites and is not transmitted to a central location. This is particularly important when dealing with sensitive or classified information. Secondly, federated learning allows satellites to collectively learn from a diverse set of data sources, benefiting from the distributed knowledge across the satellite network. Lastly, the use of federated learning reduces the communication bandwidth requirements between satellites and the central server, as only model updates are exchanged instead of raw data. By leveraging federated learning, satellites can collaborate and continuously improve their machine learning models while preserving data privacy and minimizing communication overhead. This enables the development of more intelligent and efficient satellite systems for various applications, such as Earth observation, weather forecasting, and space exploration.
☆ Integrated Sensing, Computation, and Communication for UAV-assisted Federated Edge Learning
Federated edge learning (FEEL) enables privacy-preserving model training through periodic communication between edge devices and the server. Unmanned Aerial Vehicle (UAV)-mounted edge devices are particularly advantageous for FEEL due to their flexibility and mobility in efficient data collection. In UAV-assisted FEEL, sensing, computation, and communication are coupled and compete for limited onboard resources, and UAV deployment also affects sensing and communication performance. Therefore, the joint design of UAV deployment and resource allocation is crucial to achieving the optimal training performance. In this paper, we address the problem of joint UAV deployment design and resource allocation for FEEL via a concrete case study of human motion recognition based on wireless sensing. We first analyze the impact of UAV deployment on the sensing quality and identify a threshold value for the sensing elevation angle that guarantees a satisfactory quality of data samples. Due to the non-ideal sensing channels, we consider the probabilistic sensing model, where the successful sensing probability of each UAV is determined by its position. Then, we derive the upper bound of the FEEL training loss as a function of the sensing probability. Theoretical results suggest that the convergence rate can be improved if UAVs have a uniform successful sensing probability. Based on this analysis, we formulate a training time minimization problem by jointly optimizing UAV deployment, integrated sensing, computation, and communication (ISCC) resources under a desirable optimality gap constraint. To solve this challenging mixed-integer non-convex problem, we apply the alternating optimization technique, and propose the bandwidth, batch size, and position optimization (BBPO) scheme to optimize these three decision variables alternately.
☆ Brain tumor segmentation using synthetic MR images -- A comparison of GANs and diffusion models
Large annotated datasets are required for training deep learning models, but in medical imaging data sharing is often complicated due to ethics, anonymization and data protection legislation (e.g. the general data protection regulation (GDPR)). Generative AI models, such as generative adversarial networks (GANs) and diffusion models, can today produce very realistic synthetic images, and can potentially facilitate data sharing as GDPR should not apply for medical images which do not belong to a specific person. However, in order to share synthetic images it must first be demonstrated that they can be used for training different networks with acceptable performance. Here, we therefore comprehensively evaluate four GANs (progressive GAN, StyleGAN 1-3) and a diffusion model for the task of brain tumor segmentation. Our results show that segmentation networks trained on synthetic images reach Dice scores that are 80\% - 90\% of Dice scores when training with real images, but that memorization of the training images can be a problem for diffusion models if the original dataset is too small. Furthermore, we demonstrate that common metrics for evaluating synthetic images, Fr\'echet inception distance (FID) and inception score (IS), do not correlate well with the obtained performance when using the synthetic images for training segmentation networks.
comment: 20 Pages
☆ A Deep Learning Approach Utilizing Covariance Matrix Analysis for the ISBI Edited MRS Reconstruction Challenge
This work proposes a method to accelerate the acquisition of high-quality edited magnetic resonance spectroscopy (MRS) scans using machine learning models taking the sample covariance matrix as input. The method is invariant to the number of transients and robust to noisy input data for both synthetic as well as in-vivo scenarios.
☆ Simultaneous or Sequential Training? How Speech Representations Cooperate in a Multi-Task Self-Supervised Learning System
Speech representation learning with self-supervised algorithms has resulted in notable performance boosts in many downstream tasks. Recent work combined self-supervised learning (SSL) and visually grounded speech (VGS) processing mechanisms for representation learning. The joint training with SSL and VGS mechanisms provides the opportunity to utilize both unlabeled speech and speech-related visual information based on data availability. This has shown to enhance the quality of learned representations, especially at encoding semantic- and lexical-level knowledge. In this work, we further study the joint optimization of wav2vec 2.0-based SSL and transformer-based VGS as a multi-task learning system. We explore a set of training scenarios to understand how speech representations are shared or transferred between the two tasks, and what is the optimal training strategy for cross-modal semantic retrieval and phoneme discrimination performance. As a result, we find that sequential training with wav2vec 2.0 first and VGS next provides higher performance on audio-visual retrieval compared to simultaneous optimization of both learning mechanisms. However, the parallel SSL-VGS training reduces the effects of catastrophic forgetting when switching between optimization criteria. Moreover, the results suggest that phonemic representations learned through the VGS mechanism may generalize better across datasets compared to those learned with SSL.
comment: 5 pages, accepted by EUSIPCO 2023
☆ Online Learning with Feedback Graphs: The True Shape of Regret
Sequential learning with feedback graphs is a natural extension of the multi-armed bandit problem where the problem is equipped with an underlying graph structure that provides additional information - playing an action reveals the losses of all the neighbors of the action. This problem was introduced by \citet{mannor2011} and received considerable attention in recent years. It is generally stated in the literature that the minimax regret rate for this problem is of order $\sqrt{\alpha T}$, where $\alpha$ is the independence number of the graph, and $T$ is the time horizon. However, this is proven only when the number of rounds $T$ is larger than $\alpha^3$, which poses a significant restriction for the usability of this result in large graphs. In this paper, we define a new quantity $R^*$, called the \emph{problem complexity}, and prove that the minimax regret is proportional to $R^*$ for any graph and time horizon $T$. Introducing an intricate exploration strategy, we define the \mainAlgorithm algorithm that achieves the minimax optimal regret bound and becomes the first provably optimal algorithm for this setting, even if $T$ is smaller than $\alpha^3$.
☆ Time Interpret: a Unified Model Interpretability Library for Time Series
We introduce $\texttt{time_interpret}$, a library designed as an extension of Captum, with a specific focus on temporal data. As such, this library implements several feature attribution methods that can be used to explain predictions made by any Pytorch model. $\texttt{time_interpret}$ also provides several synthetic and real world time series datasets, various PyTorch models, as well as a set of methods to evaluate feature attributions. Moreover, while being primarily developed to explain predictions based on temporal data, some of its components have a different application, including for instance methods explaining predictions made by language models. In this paper, we give a general introduction of this library. We also present several previously unpublished feature attribution methods, which have been developed along with $\texttt{time_interpret}$.
comment: 7 pages, 1 figure. Code available at https://github.com/josephenguehard/time_interpret
☆ Best of Both Worlds: Hybrid SNN-ANN Architecture for Event-based Optical Flow Estimation
Event-based cameras offer a low-power alternative to frame-based cameras for capturing high-speed motion and high dynamic range scenes. They provide asynchronous streams of sparse events. Spiking Neural Networks (SNNs) with their asynchronous event-driven compute, show great potential for extracting the spatio-temporal features from these event streams. In contrast, the standard Analog Neural Networks (ANNs1) fail to process event data effectively. However, training SNNs is difficult due to additional trainable parameters (thresholds and leaks), vanishing spikes at deeper layers, non-differentiable binary activation function etc. Moreover, an additional data structure "membrane potential" responsible for keeping track of temporal information, must be fetched and updated at every timestep in SNNs. To overcome these, we propose a novel SNN-ANN hybrid architecture that combines the strengths of both. Specifically, we leverage the asynchronous compute capabilities of SNN layers to effectively extract the input temporal information. While the ANN layers offer trouble-free training and implementation on standard machine learning hardware such as GPUs. We provide extensive experimental analysis for assigning each layer to be spiking or analog in nature, leading to a network configuration optimized for performance and ease of training. We evaluate our hybrid architectures for optical flow estimation using event-data on DSEC-flow and Mutli-Vehicle Stereo Event-Camera (MVSEC) datasets. The results indicate that our configured hybrid architectures outperform the state-of-the-art ANN-only, SNN-only and past hybrid architectures both in terms of accuracy and efficiency. Specifically, our hybrid architecture exhibit a 31% and 24.8% lower average endpoint error (AEE) at 2.1x and 3.1x lower energy, compared to an SNN-only architecture on DSEC and MVSEC datasets, respectively.
☆ Complex Preferences for Different Convergent Priors in Discrete Graph Diffusion
Diffusion models have achieved state-of-the-art performance in generating many different kinds of data, including images, text, and videos. Despite their success, there has been limited research on how the underlying diffusion process and the final convergent prior can affect generative performance; this research has also been limited to continuous data types and a score-based diffusion framework. To fill this gap, we explore how different discrete diffusion kernels (which converge to different prior distributions) affect the performance of diffusion models for graphs. To this end, we developed a novel formulation of a family of discrete diffusion kernels which are easily adjustable to converge to different Bernoulli priors, and we study the effect of these different kernels on generative performance. We show that the quality of generated graphs is sensitive to the prior used, and that the optimal choice cannot be explained by obvious statistics or metrics, which challenges the intuitions which previous works have suggested.
☆ A Simple and Flexible Modeling for Mental Disorder Detection by Learning from Clinical Questionnaires ACL 2023
Social media is one of the most highly sought resources for analyzing characteristics of the language by its users. In particular, many researchers utilized various linguistic features of mental health problems from social media. However, existing approaches to detecting mental disorders face critical challenges, such as the scarcity of high-quality data or the trade-off between addressing the complexity of models and presenting interpretable results grounded in expert domain knowledge. To address these challenges, we design a simple but flexible model that preserves domain-based interpretability. We propose a novel approach that captures the semantic meanings directly from the text and compares them to symptom-related descriptions. Experimental results demonstrate that our model outperforms relevant baselines on various mental disorder detection tasks. Our detailed analysis shows that the proposed model is effective at leveraging domain knowledge, transferable to other mental disorders, and providing interpretable detection results.
comment: ACL 2023, 15 pages, 11 tables, 4 figures
☆ Random Distribution Shift in Refugee Placement: Strategies for Building Robust Models
Algorithmic assignment of refugees and asylum seekers to locations within host countries has gained attention in recent years, with implementations in the US and Switzerland. These approaches use data on past arrivals to generate machine learning models that can be used (along with assignment algorithms) to match families to locations, with the goal of maximizing a policy-relevant integration outcome such as employment status after a certain duration. Existing implementations and research train models to predict the policy outcome directly, and use these predictions in the assignment procedure. However, the merits of this approach, particularly in non-stationary settings, has not been previously explored. This study proposes and compares three different modeling strategies: the standard approach described above, an approach that uses newer data and proxy outcomes, and a hybrid approach. We show that the hybrid approach is robust to both distribution shift and weak proxy relationships -- the failure points of the other two methods, respectively. We compare these approaches empirically using data on asylum seekers in the Netherlands. Surprisingly, we find that both the proxy and hybrid approaches out-perform the standard approach in practice. These insights support the development of a real-world recommendation tool currently used by NGOs and government agencies.
☆ Continual Learning with Pretrained Backbones by Tuning in the Input Space
The intrinsic difficulty in adapting deep learning models to non-stationary environments limits the applicability of neural networks to real-world tasks. This issue is critical in practical supervised learning settings, such as the ones in which a pre-trained model computes projections toward a latent space where different task predictors are sequentially learned over time. As a matter of fact, incrementally fine-tuning the whole model to better adapt to new tasks usually results in catastrophic forgetting, with decreasing performance over the past experiences and losing valuable knowledge from the pre-training stage. In this paper, we propose a novel strategy to make the fine-tuning procedure more effective, by avoiding to update the pre-trained part of the network and learning not only the usual classification head, but also a set of newly-introduced learnable parameters that are responsible for transforming the input data. This process allows the network to effectively leverage the pre-training knowledge and find a good trade-off between plasticity and stability with modest computational efforts, thus especially suitable for on-the-edge settings. Our experiments on four image classification problems in a continual learning setting confirm the quality of the proposed approach when compared to several fine-tuning procedures and to popular continual learning methods.
☆ Improved Stability and Generalization Analysis of the Decentralized SGD Algorithm
This paper presents a new generalization error analysis for the Decentralized Stochastic Gradient Descent (D-SGD) algorithm based on algorithmic stability. The obtained results largely improve upon state-of-the-art results, and even invalidate their claims that the communication graph has a detrimental effect on generalization. For instance, we show that in convex settings, D-SGD has the same generalization bounds as the classical SGD algorithm, no matter the choice of graph. We exhibit that this counter-intuitive result comes from considering the average of local parameters, which hides a final global averaging step incompatible with the decentralized scenario. In light of this observation, we advocate to analyze the supremum over local parameters and show that in this case, the graph does have an impact on the generalization. Unlike prior results, our analysis yields non-vacuous bounds even for non-connected graphs.
☆ Causal Discovery using Bayesian Model Selection
With only observational data on two variables, and without other assumptions, it is not possible to infer which one causes the other. Much of the causal literature has focused on guaranteeing identifiability of causal direction in statistical models for datasets where strong assumptions hold, such as additive noise or restrictions on parameter count. These methods are then subsequently tested on realistic datasets, most of which violate their assumptions. Building on previous attempts, we show how to use causal assumptions within the Bayesian framework. This allows us to specify models with realistic assumptions, while also encoding independent causal mechanisms, leading to an asymmetry between the causal directions. Identifying causal direction then becomes a Bayesian model selection problem. We analyse why Bayesian model selection works for known identifiable cases and flexible model classes, while also providing correctness guarantees about its behaviour. To demonstrate our approach, we construct a Bayesian non-parametric model that can flexibly model the joint. We then outperform previous methods on a wide range of benchmark datasets with varying data generating assumptions showing the usefulness of our method.
☆ Decentralized SGD and Average-direction SAM are Asymptotically Equivalent ICML 2023
Decentralized stochastic gradient descent (D-SGD) allows collaborative learning on massive devices simultaneously without the control of a central server. However, existing theories claim that decentralization invariably undermines generalization. In this paper, we challenge the conventional belief and present a completely new perspective for understanding decentralized learning. We prove that D-SGD implicitly minimizes the loss function of an average-direction Sharpness-aware minimization (SAM) algorithm under general non-convex non-$\beta$-smooth settings. This surprising asymptotic equivalence reveals an intrinsic regularization-optimization trade-off and three advantages of decentralization: (1) there exists a free uncertainty evaluation mechanism in D-SGD to improve posterior estimation; (2) D-SGD exhibits a gradient smoothing effect; and (3) the sharpness regularization effect of D-SGD does not decrease as total batch size increases, which justifies the potential generalization benefit of D-SGD over centralized SGD (C-SGD) in large-batch scenarios.
comment: Accepted for publication in the 40th International Conference on Machine Learning (ICML 2023)
☆ Learning nonparametric latent causal graphs with unknown interventions
We establish conditions under which latent causal graphs are nonparametrically identifiable and can be reconstructed from unknown interventions in the latent space. Our primary focus is the identification of the latent structure in a measurement model, i.e. causal graphical models where dependence between observed variables is insignificant compared to dependence between latent representations, without making parametric assumptions such as linearity or Gaussianity. Moreover, we do not assume the number of hidden variables is known, and we show that at most one unknown intervention per hidden variable is needed. This extends a recent line of work on learning causal representations from observations and interventions. The proofs are constructive and introduce two new graphical concepts -- imaginary subsets and isolated edges -- that may be useful in their own right. As a matter of independent interest, the proofs also involve a novel characterization of the limits of edge orientations within the equivalence class of DAGs induced by unknown interventions. Experiments confirm that the latent graph can be recovered from data using our theoretical results. These are the first results to characterize the conditions under which causal representations are identifiable without making any parametric assumptions in a general setting with unknown interventions and without faithfulness.
☆ Representational Strengths and Limitations of Transformers
Attention layers, as commonly used in transformers, form the backbone of modern deep learning, yet there is no mathematical description of their benefits and deficiencies as compared with other architectures. In this work we establish both positive and negative results on the representation power of attention layers, with a focus on intrinsic complexity parameters such as width, depth, and embedding dimension. On the positive side, we present a sparse averaging task, where recurrent networks and feedforward networks all have complexity scaling polynomially in the input size, whereas transformers scale merely logarithmically in the input size; furthermore, we use the same construction to show the necessity and role of a large embedding dimension in a transformer. On the negative side, we present a triple detection task, where attention layers in turn have complexity scaling linearly in the input size; as this scenario seems rare in practice, we also present natural variants that can be efficiently solved by attention layers. The proof techniques emphasize the value of communication complexity in the analysis of transformers and related models, and the role of sparse averaging as a prototypical attention task, which even finds use in the analysis of triple detection.
☆ Evading Black-box Classifiers Without Breaking Eggs
Decision-based evasion attacks repeatedly query a black-box classifier to generate adversarial examples. Prior work measures the cost of such attacks by the total number of queries made to the classifier. We argue this metric is flawed. Most security-critical machine learning systems aim to weed out "bad" data (e.g., malware, harmful content, etc). Queries to such systems carry a fundamentally asymmetric cost: queries detected as "bad" come at a higher cost because they trigger additional security filters, e.g., usage throttling or account suspension. Yet, we find that existing decision-based attacks issue a large number of "bad" queries, which likely renders them ineffective against security-critical systems. We then design new attacks that reduce the number of bad queries by $1.5$-$7.3\times$, but often at a significant increase in total (non-bad) queries. We thus pose it as an open problem to build black-box attacks that are more effective under realistic cost metrics.
comment: Code at https://github.com/ethz-privsec/realistic-adv-examples
☆ Image Reconstruction for Accelerated MR Scan with Faster Fourier Convolutional Neural Networks
Partial scan is a common approach to accelerate Magnetic Resonance Imaging (MRI) data acquisition in both 2D and 3D settings. However, accurately reconstructing images from partial scan data (i.e., incomplete k-space matrices) remains challenging due to lack of an effectively global receptive field in both spatial and k-space domains. To address this problem, we propose the following: (1) a novel convolutional operator called Faster Fourier Convolution (FasterFC) to replace the two consecutive convolution operations typically used in convolutional neural networks (e.g., U-Net, ResNet). Based on the spectral convolution theorem in Fourier theory, FasterFC employs alternating kernels of size 1 in 3D case) in different domains to extend the dual-domain receptive field to the global and achieves faster calculation speed than traditional Fast Fourier Convolution (FFC). (2) A 2D accelerated MRI method, FasterFC-End-to-End-VarNet, which uses FasterFC to improve the sensitivity maps and reconstruction quality. (3) A multi-stage 3D accelerated MRI method called FasterFC-based Single-to-group Network (FAS-Net) that utilizes a single-to-group algorithm to guide k-space domain reconstruction, followed by FasterFC-based cascaded convolutional neural networks to expand the effective receptive field in the dual-domain. Experimental results on the fastMRI and Stanford MRI Data datasets demonstrate that FasterFC improves the quality of both 2D and 3D reconstruction. Moreover, FAS-Net, as a 3D high-resolution multi-coil (eight) accelerated MRI method, achieves superior reconstruction performance in both qualitative and quantitative results compared with state-of-the-art 2D and 3D methods.
☆ Neuron Activation Coverage: Rethinking Out-of-distribution Detection and Generalization
The out-of-distribution (OOD) problem generally arises when neural networks encounter data that significantly deviates from the training data distribution, \ie, in-distribution (InD). In this paper, we study the OOD problem from a neuron activation view. We first formulate neuron activation states by considering both the neuron output and its influence on model decisions. Then, we propose the concept of \textit{neuron activation coverage} (NAC), which characterizes the neuron behaviors under InD and OOD data. Leveraging our NAC, we show that 1) InD and OOD inputs can be naturally separated based on the neuron behavior, which significantly eases the OOD detection problem and achieves a record-breaking performance of 0.03% FPR95 on ResNet-50, outperforming the previous best method by 20.67%; 2) a positive correlation between NAC and model generalization ability consistently holds across architectures and datasets, which enables a NAC-based criterion for evaluating model robustness. By comparison with the traditional validation criterion, we show that NAC-based criterion not only can select more robust models, but also has a stronger correlation with OOD test performance.
☆ Data-Driven Regret Balancing for Online Model Selection in Bandits
We consider model selection for sequential decision making in stochastic environments with bandit feedback, where a meta-learner has at its disposal a pool of base learners, and decides on the fly which action to take based on the policies recommended by each base learner. Model selection is performed by regret balancing but, unlike the recent literature on this subject, we do not assume any prior knowledge about the base learners like candidate regret guarantees; instead, we uncover these quantities in a data-driven manner. The meta-learner is therefore able to leverage the realized regret incurred by each base learner for the learning environment at hand (as opposed to the expected regret), and single out the best such regret. We design two model selection algorithms operating with this more ambitious notion of regret and, besides proving model selection guarantees via regret balancing, we experimentally demonstrate the compelling practical benefits of dealing with actual regrets instead of candidate regret bounds.
☆ Learning Probabilistic Symmetrization for Architecture Agnostic Equivariance
We present a novel framework to overcome the limitations of equivariant architectures in learning functions with group symmetries. In contrary to equivariant architectures, we use an arbitrary base model (such as an MLP or a transformer) and symmetrize it to be equivariant to the given group by employing a small equivariant network that parameterizes the probabilistic distribution underlying the symmetrization. The distribution is end-to-end trained with the base model which can maximize performance while reducing sample complexity of symmetrization. We show that this approach ensures not only equivariance to given group but also universal approximation capability in expectation. We implement our method on a simple patch-based transformer that can be initialized from pretrained vision transformers, and test it for a wide range of symmetry groups including permutation and Euclidean groups and their combinations. Empirical tests show competitive results against tailored equivariant architectures, suggesting the potential for learning equivariant functions for diverse groups using a non-equivariant universal base architecture. We further show evidence of enhanced learning in symmetric modalities, like graphs, when pretrained from non-symmetric modalities, like vision. Our implementation will be open-sourced at https://github.com/jw9730/lps.
comment: 25 pages, 3 figures
☆ Seizing Serendipity: Exploiting the Value of Past Success in Off-Policy Actor-Critic
Learning high-quality Q-value functions plays a key role in the success of many modern off-policy deep reinforcement learning (RL) algorithms. Previous works focus on addressing the value overestimation issue, an outcome of adopting function approximators and off-policy learning. Deviating from the common viewpoint, we observe that Q-values are indeed underestimated in the latter stage of the RL training process, primarily related to the use of inferior actions from the current policy in Bellman updates as compared to the more optimal action samples in the replay buffer. We hypothesize that this long-neglected phenomenon potentially hinders policy learning and reduces sample efficiency. Our insight to address this issue is to incorporate sufficient exploitation of past successes while maintaining exploration optimism. We propose the Blended Exploitation and Exploration (BEE) operator, a simple yet effective approach that updates Q-value using both historical best-performing actions and the current policy. The instantiations of our method in both model-free and model-based settings outperform state-of-the-art methods in various continuous control tasks and achieve strong performance in failure-prone scenarios and real-world robot tasks.
☆ Local Boosting for Weakly-Supervised Learning KDD 2023
Boosting is a commonly used technique to enhance the performance of a set of base models by combining them into a strong ensemble model. Though widely adopted, boosting is typically used in supervised learning where the data is labeled accurately. However, in weakly supervised learning, where most of the data is labeled through weak and noisy sources, it remains nontrivial to design effective boosting approaches. In this work, we show that the standard implementation of the convex combination of base learners can hardly work due to the presence of noisy labels. Instead, we propose $\textit{LocalBoost}$, a novel framework for weakly-supervised boosting. LocalBoost iteratively boosts the ensemble model from two dimensions, i.e., intra-source and inter-source. The intra-source boosting introduces locality to the base learners and enables each base learner to focus on a particular feature regime by training new base learners on granularity-varying error regions. For the inter-source boosting, we leverage a conditional function to indicate the weak source where the sample is more likely to appear. To account for the weak labels, we further design an estimate-then-modify approach to compute the model weights. Experiments on seven datasets show that our method significantly outperforms vanilla boosting methods and other weakly-supervised methods.
comment: Accepted by KDD 2023 Research Track
☆ HireVAE: An Online and Adaptive Factor Model Based on Hierarchical and Regime-Switch VAE IJCAI 2023
Factor model is a fundamental investment tool in quantitative investment, which can be empowered by deep learning to become more flexible and efficient in practical complicated investing situations. However, it is still an open question to build a factor model that can conduct stock prediction in an online and adaptive setting, where the model can adapt itself to match the current market regime identified based on only point-in-time market information. To tackle this problem, we propose the first deep learning based online and adaptive factor model, HireVAE, at the core of which is a hierarchical latent space that embeds the underlying relationship between the market situation and stock-wise latent factors, so that HireVAE can effectively estimate useful latent factors given only historical market information and subsequently predict accurate stock returns. Across four commonly used real stock market benchmarks, the proposed HireVAE demonstrate superior performance in terms of active returns over previous methods, verifying the potential of such online and adaptive factor model.
comment: Accepted to IJCAI 2023
☆ Learning to Substitute Spans towards Improving Compositional Generalization ACL 2023
Despite the rising prevalence of neural sequence models, recent empirical evidences suggest their deficiency in compositional generalization. One of the current de-facto solutions to this problem is compositional data augmentation, aiming to incur additional compositional inductive bias. Nonetheless, the improvement offered by existing handcrafted augmentation strategies is limited when successful systematic generalization of neural sequence models requires multi-grained compositional bias (i.e., not limited to either lexical or structural biases only) or differentiation of training sequences in an imbalanced difficulty distribution. To address the two challenges, we first propose a novel compositional augmentation strategy dubbed \textbf{Span} \textbf{Sub}stitution (SpanSub) that enables multi-grained composition of substantial substructures in the whole training set. Over and above that, we introduce the \textbf{L}earning \textbf{to} \textbf{S}ubstitute \textbf{S}pan (L2S2) framework which empowers the learning of span substitution probabilities in SpanSub in an end-to-end manner by maximizing the loss of neural sequence models, so as to outweigh those challenging compositions with elusive concepts and novel surroundings. Our empirical results on three standard compositional generalization benchmarks, including SCAN, COGS and GeoQuery (with an improvement of at most 66.5\%, 10.3\%, 1.2\%, respectively), demonstrate the superiority of SpanSub, %the learning framework L2S2 and their combination.
comment: accepted by ACL 2023
☆ Computational Complexity of Detecting Proximity to Losslessly Compressible Neural Network Parameters
To better understand complexity in neural networks, we theoretically investigate the idealised phenomenon of lossless network compressibility, whereby an identical function can be implemented with a smaller network. We give an efficient formal algorithm for optimal lossless compression in the setting of single-hidden-layer hyperbolic tangent networks. To measure lossless compressibility, we define the rank of a parameter as the minimum number of hidden units required to implement the same function. Losslessly compressible parameters are atypical, but their existence has implications for nearby parameters. We define the proximate rank of a parameter as the rank of the most compressible parameter within a small $L^\infty$ neighbourhood. Unfortunately, detecting nearby losslessly compressible parameters is not so easy: we show that bounding the proximate rank is an NP-complete problem, using a reduction from Boolean satisfiability via a geometric problem involving covering points in the plane with small squares. These results underscore the computational complexity of measuring neural network complexity, laying a foundation for future theoretical and empirical work in this direction.
comment: 9 pages paper, 31 pages total, 9 figures, 3 tables
☆ The $L^\infty$ Learnability of Reproducing Kernel Hilbert Spaces
In this work, we analyze the learnability of reproducing kernel Hilbert spaces (RKHS) under the $L^\infty$ norm, which is critical for understanding the performance of kernel methods and random feature models in safety- and security-critical applications. Specifically, we relate the $L^\infty$ learnability of a RKHS to the spectrum decay of the associate kernel and both lower bounds and upper bounds of the sample complexity are established. In particular, for dot-product kernels on the sphere, we identify conditions when the $L^\infty$ learning can be achieved with polynomial samples. Let $d$ denote the input dimension and assume the kernel spectrum roughly decays as $\lambda_k\sim k^{-1-\beta}$ with $\beta>0$. We prove that if $\beta$ is independent of the input dimension $d$, then functions in the RKHS can be learned efficiently under the $L^\infty$ norm, i.e., the sample complexity depends polynomially on $d$. In contrast, if $\beta=1/\mathrm{poly}(d)$, then the $L^\infty$ learning requires exponentially many samples.
comment: 20 pages
☆ MM-DAG: Multi-task DAG Learning for Multi-modal Data -- with Application for Traffic Congestion Analysis KDD 2023
This paper proposes to learn Multi-task, Multi-modal Direct Acyclic Graphs (MM-DAGs), which are commonly observed in complex systems, e.g., traffic, manufacturing, and weather systems, whose variables are multi-modal with scalars, vectors, and functions. This paper takes the traffic congestion analysis as a concrete case, where a traffic intersection is usually regarded as a DAG. In a road network of multiple intersections, different intersections can only have some overlapping and distinct variables observed. For example, a signalized intersection has traffic light-related variables, whereas unsignalized ones do not. This encourages the multi-task design: with each DAG as a task, the MM-DAG tries to learn the multiple DAGs jointly so that their consensus and consistency are maximized. To this end, we innovatively propose a multi-modal regression for linear causal relationship description of different variables. Then we develop a novel Causality Difference (CD) measure and its differentiable approximator. Compared with existing SOTA measures, CD can penalize the causal structural difference among DAGs with distinct nodes and can better consider the uncertainty of causal orders. We rigidly prove our design's topological interpretation and consistency properties. We conduct thorough simulations and one case study to show the effectiveness of our MM-DAG. The code is available under https://github.com/Lantian72/MM-DAG
comment: Accepted in SIGKDD 2023
☆ Near-Optimal Quantum Coreset Construction Algorithms for Clustering ICML 2023
$k$-Clustering in $\mathbb{R}^d$ (e.g., $k$-median and $k$-means) is a fundamental machine learning problem. While near-linear time approximation algorithms were known in the classical setting for a dataset with cardinality $n$, it remains open to find sublinear-time quantum algorithms. We give quantum algorithms that find coresets for $k$-clustering in $\mathbb{R}^d$ with $\tilde{O}(\sqrt{nk}d^{3/2})$ query complexity. Our coreset reduces the input size from $n$ to $\mathrm{poly}(k\epsilon^{-1}d)$, so that existing $\alpha$-approximation algorithms for clustering can run on top of it and yield $(1 + \epsilon)\alpha$-approximation. This eventually yields a quadratic speedup for various $k$-clustering approximation algorithms. We complement our algorithm with a nearly matching lower bound, that any quantum algorithm must make $\Omega(\sqrt{nk})$ queries in order to achieve even $O(1)$-approximation for $k$-clustering.
comment: Comments: 32 pages, 0 figures, 1 table. To appear in the Fortieth International Conference on Machine Learning (ICML 2023)
☆ COMET: Learning Cardinality Constrained Mixture of Experts with Trees and Local Search KDD 2023
The sparse Mixture-of-Experts (Sparse-MoE) framework efficiently scales up model capacity in various domains, such as natural language processing and vision. Sparse-MoEs select a subset of the "experts" (thus, only a portion of the overall network) for each input sample using a sparse, trainable gate. Existing sparse gates are prone to convergence and performance issues when training with first-order optimization methods. In this paper, we introduce two improvements to current MoE approaches. First, we propose a new sparse gate: COMET, which relies on a novel tree-based mechanism. COMET is differentiable, can exploit sparsity to speed up computation, and outperforms state-of-the-art gates. Second, due to the challenging combinatorial nature of sparse expert selection, first-order methods are typically prone to low-quality solutions. To deal with this challenge, we propose a novel, permutation-based local search method that can complement first-order methods in training any sparse gate, e.g., Hash routing, Top-k, DSelect-k, and COMET. We show that local search can help networks escape bad initializations or solutions. We performed large-scale experiments on various domains, including recommender systems, vision, and natural language processing. On standard vision and recommender systems benchmarks, COMET+ (COMET with local search) achieves up to 13% improvement in ROC AUC over popular gates, e.g., Hash routing and Top-k, and up to 9% over prior differentiable gates e.g., DSelect-k. When Top-k and Hash gates are combined with local search, we see up to $100\times$ reduction in the budget needed for hyperparameter tuning. Moreover, for language modeling, our approach improves over the state-of-the-art MoEBERT model for distilling BERT on 5/7 GLUE benchmarks as well as SQuAD dataset.
comment: Accepted in KDD 2023
☆ Discovering Dynamic Causal Space for DAG Structure Learning KDD 2023
Discovering causal structure from purely observational data (i.e., causal discovery), aiming to identify causal relationships among variables, is a fundamental task in machine learning. The recent invention of differentiable score-based DAG learners is a crucial enabler, which reframes the combinatorial optimization problem into a differentiable optimization with a DAG constraint over directed graph space. Despite their great success, these cutting-edge DAG learners incorporate DAG-ness independent score functions to evaluate the directed graph candidates, lacking in considering graph structure. As a result, measuring the data fitness alone regardless of DAG-ness inevitably leads to discovering suboptimal DAGs and model vulnerabilities. Towards this end, we propose a dynamic causal space for DAG structure learning, coined CASPER, that integrates the graph structure into the score function as a new measure in the causal space to faithfully reflect the causal distance between estimated and ground truth DAG. CASPER revises the learning process as well as enhances the DAG structure learning via adaptive attention to DAG-ness. Grounded by empirical visualization, CASPER, as a space, satisfies a series of desired properties, such as structure awareness and noise robustness. Extensive experiments on both synthetic and real-world datasets clearly validate the superiority of our CASPER over the state-of-the-art causal discovery methods in terms of accuracy and robustness.
comment: Accepted by KDD 2023
☆ MultiAdam: Parameter-wise Scale-invariant Optimizer for Multiscale Training of Physics-informed Neural Networks
Physics-informed Neural Networks (PINNs) have recently achieved remarkable progress in solving Partial Differential Equations (PDEs) in various fields by minimizing a weighted sum of PDE loss and boundary loss. However, there are several critical challenges in the training of PINNs, including the lack of theoretical frameworks and the imbalance between PDE loss and boundary loss. In this paper, we present an analysis of second-order non-homogeneous PDEs, which are classified into three categories and applicable to various common problems. We also characterize the connections between the training loss and actual error, guaranteeing convergence under mild conditions. The theoretical analysis inspires us to further propose MultiAdam, a scale-invariant optimizer that leverages gradient momentum to parameter-wisely balance the loss terms. Extensive experiment results on multiple problems from different physical domains demonstrate that our MultiAdam solver can improve the predictive accuracy by 1-2 orders of magnitude compared with strong baselines.
Transformer-Based UNet with Multi-Headed Cross-Attention Skip Connections to Eliminate Artifacts in Scanned Documents
The extraction of text in high quality is essential for text-based document analysis tasks like Document Classification or Named Entity Recognition. Unfortunately, this is not always ensured, as poor scan quality and the resulting artifacts lead to errors in the Optical Character Recognition (OCR) process. Current approaches using Convolutional Neural Networks show promising results for background removal tasks but fail correcting artifacts like pixelation or compression errors. For general images, Transformer backbones are getting integrated more frequently in well-known neural network structures for denoising tasks. In this work, a modified UNet structure using a Swin Transformer backbone is presented to remove typical artifacts in scanned documents. Multi-headed cross-attention skip connections are used to more selectively learn features in respective levels of abstraction. The performance of this approach is examined regarding compression errors, pixelation and random noise. An improvement in text extraction quality with a reduced error rate of up to 53.9% on the synthetic data is archived. The pretrained base-model can be easily adapted to new artifacts. The cross-attention skip connections allow to integrate textual information extracted from the encoder or in form of commands to more selectively control the models outcome. The latter is shown by means of an example application.
☆ Deep Active Learning with Structured Neural Depth Search
Previous work optimizes traditional active learning (AL) processes with incremental neural network architecture search (Active-iNAS) based on data complexity change, which improves the accuracy and learning efficiency. However, Active-iNAS trains several models and selects the model with the best generalization performance for querying the subsequent samples after each active learning cycle. The independent training processes lead to an insufferable computational budget, which is significantly inefficient and limits search flexibility and final performance. To address this issue, we propose a novel active strategy with the method called structured variational inference (SVI) or structured neural depth search (SNDS) whereby we could use the gradient descent method in neural network depth search during AL processes. At the same time, we theoretically demonstrate that the current VI-based methods based on the mean-field assumption could lead to poor performance. We apply our strategy using three querying techniques and three datasets and show that our strategy outperforms current methods.
comment: 10 pages, 8 figures, prepare for TNNLS
☆ On Tail Decay Rate Estimation of Loss Function Distributions
The study of loss function distributions is critical to characterize a model's behaviour on a given machine learning problem. For example, while the quality of a model is commonly determined by the average loss assessed on a testing set, this quantity does not reflect the existence of the true mean of the loss distribution. Indeed, the finiteness of the statistical moments of the loss distribution is related to the thickness of its tails, which are generally unknown. Since typical cross-validation schemes determine a family of testing loss distributions conditioned on the training samples, the total loss distribution must be recovered by marginalizing over the space of training sets. As we show in this work, the finiteness of the sampling procedure negatively affects the reliability and efficiency of classical tail estimation methods from the Extreme Value Theory, such as the Peaks-Over-Threshold approach. In this work we tackle this issue by developing a novel general theory for estimating the tails of marginal distributions, when there exists a large variability between locations of the individual conditional distributions underlying the marginal. To this end, we demonstrate that under some regularity conditions, the shape parameter of the marginal distribution is the maximum tail shape parameter of the family of conditional distributions. We term this estimation approach as Cross Tail Estimation (CTE). We test cross-tail estimation in a series of experiments on simulated and real data, showing the improved robustness and quality of tail estimation as compared to classical approaches, and providing evidence for the relationship between overfitting and loss distribution tail thickness.
☆ A Data-driven Region Generation Framework for Spatiotemporal Transportation Service Management
MAUP (modifiable areal unit problem) is a fundamental problem for spatial data management and analysis. As an instantiation of MAUP in online transportation platforms, region generation (i.e., specifying the areal unit for service operations) is the first and vital step for supporting spatiotemporal transportation services such as ride-sharing and freight transport. Most existing region generation methods are manually specified (e.g., fixed-size grids), suffering from poor spatial semantic meaning and inflexibility to meet service operation requirements. In this paper, we propose RegionGen, a data-driven region generation framework that can specify regions with key characteristics (e.g., good spatial semantic meaning and predictability) by modeling region generation as a multi-objective optimization problem. First, to obtain good spatial semantic meaning, RegionGen segments the whole city into atomic spatial elements based on road networks and obstacles (e.g., rivers). Then, it clusters the atomic spatial elements into regions by maximizing various operation characteristics, which is formulated as a multi-objective optimization problem. For this optimization problem, we propose a multi-objective co-optimization algorithm. Extensive experiments verify that RegionGen can generate more suitable regions than traditional methods for spatiotemporal service management.
☆ Using Multiple Dermoscopic Photographs of One Lesion Improves Melanoma Classification via Deep Learning: A Prognostic Diagnostic Accuracy Study
Background: Convolutional neural network (CNN)-based melanoma classifiers face several challenges that limit their usefulness in clinical practice. Objective: To investigate the impact of multiple real-world dermoscopic views of a single lesion of interest on a CNN-based melanoma classifier. Methods: This study evaluated 656 suspected melanoma lesions. Classifier performance was measured using area under the receiver operating characteristic curve (AUROC), expected calibration error (ECE) and maximum confidence change (MCC) for (I) a single-view scenario, (II) a multiview scenario using multiple artificially modified images per lesion and (III) a multiview scenario with multiple real-world images per lesion. Results: The multiview approach with real-world images significantly increased the AUROC from 0.905 (95% CI, 0.879-0.929) in the single-view approach to 0.930 (95% CI, 0.909-0.951). ECE and MCC also improved significantly from 0.131 (95% CI, 0.105-0.159) to 0.072 (95% CI: 0.052-0.093) and from 0.149 (95% CI, 0.125-0.171) to 0.115 (95% CI: 0.099-0.131), respectively. Comparing multiview real-world to artificially modified images showed comparable diagnostic accuracy and uncertainty estimation, but significantly worse robustness for the latter. Conclusion: Using multiple real-world images is an inexpensive method to positively impact the performance of a CNN-based melanoma classifier.
☆ Enhancing naive classifier for positive unlabeled data based on logistic regression approach
We argue that for analysis of Positive Unlabeled (PU) data under Selected Completely At Random (SCAR) assumption it is fruitful to view the problem as fitting of misspecified model to the data. Namely, we show that the results on misspecified fit imply that in the case when posterior probability of the response is modelled by logistic regression, fitting the logistic regression to the observable PU data which {\it does not} follow this model, still yields the vector of estimated parameters approximately colinear with the true vector of parameters. This observation together with choosing the intercept of the classifier based on optimisation of analogue of F1 measure yields a classifier which performs on par or better than its competitors on several real data sets considered.
☆ Modeling Human-like Concept Learning with Bayesian Inference over Natural Language
We model learning of abstract symbolic concepts by performing Bayesian inference over utterances in natural language. For efficient inference, we use a large language model as a proposal distribution. We fit a prior to human data to better model human learners, and evaluate on both generative and logical concepts.
☆ Navigating Explanatory Multiverse Through Counterfactual Path Geometry
Counterfactual explanations are the de facto standard when tasked with interpreting decisions of (opaque) predictive models. Their generation is often subject to algorithmic and domain-specific constraints -- such as density-based feasibility for the former and attribute (im)mutability or directionality of change for the latter -- that aim to maximise their real-life utility. In addition to desiderata with respect to the counterfactual instance itself, the existence of a viable path connecting it with the factual data point, known as algorithmic recourse, has become an important technical consideration. While both of these requirements ensure that the steps of the journey as well as its destination are admissible, current literature neglects the multiplicity of such counterfactual paths. To address this shortcoming we introduce the novel concept of explanatory multiverse that encompasses all the possible counterfactual journeys and shows how to navigate, reason about and compare the geometry of these paths -- their affinity, branching, divergence and possible future convergence -- with two methods: vector spaces and graphs. Implementing this (interactive) explanatory process grants explainees more agency by allowing them to select counterfactuals based on the properties of the journey leading to them in addition to their absolute differences.
☆ A survey of Generative AI Applications
Generative AI has experienced remarkable growth in recent years, leading to a wide array of applications across diverse domains. In this paper, we present a comprehensive survey of more than 350 generative AI applications, providing a structured taxonomy and concise descriptions of various unimodal and even multimodal generative AIs. The survey is organized into sections, covering a wide range of unimodal generative AI applications such as text, images, video, gaming and brain information. Our survey aims to serve as a valuable resource for researchers and practitioners to navigate the rapidly expanding landscape of generative AI, facilitating a better understanding of the current state-of-the-art and fostering further innovation in the field.
☆ Input gradient diversity for neural network ensembles
Deep Ensembles (DEs) demonstrate improved accuracy, calibration and robustness to perturbations over single neural networks partly due to their functional diversity. Particle-based variational inference (ParVI) methods enhance diversity by formalizing a repulsion term based on a network similarity kernel. However, weight-space repulsion is inefficient due to over-parameterization, while direct function-space repulsion has been found to produce little improvement over DEs. To sidestep these difficulties, we propose First-order Repulsive Deep Ensemble (FoRDE), an ensemble learning method based on ParVI, which performs repulsion in the space of first-order input gradients. As input gradients uniquely characterize a function up to translation and are much smaller in dimension than the weights, this method guarantees that ensemble members are functionally different. Intuitively, diversifying the input gradients encourages each network to learn different features, which is expected to improve the robustness of an ensemble. Experiments on image classification datasets show that FoRDE significantly outperforms the gold-standard DEs and other ensemble methods in accuracy and calibration under covariate shift due to input perturbations.
comment: Under review
☆ Identifying the style by a qualified reader on a short fragment of generated poetry
Style is an important concept in today's challenges in natural language generating. After the success in the field of image style transfer, the task of text style transfer became actual and attractive. Researchers are also interested in the tasks of style reproducing in generation of the poetic text. Evaluation of style reproducing in natural poetry generation remains a problem. I used 3 character-based LSTM-models to work with style reproducing assessment. All three models were trained on the corpus of texts by famous Russian-speaking poets. Samples were shown to the assessors and 4 answer options were offered, the style of which poet this sample reproduces. In addition, the assessors were asked how well they were familiar with the work of the poet they had named. Students studying history of literature were the assessors, 94 answers were received. It has appeared that accuracy of definition of style increases if the assessor can quote the poet by heart. Each model showed at least 0.7 macro-average accuracy. The experiment showed that it is better to involve a professional rather than a naive reader in the evaluation of style in the tasks of poetry generation, while lstm models are good at reproducing the style of Russian poets even on a limited training corpus.
comment: 6 pages, 2 tables
☆ Networked Communication for Decentralised Agents in Mean-Field Games
We introduce networked communication to the mean-field game framework. In particular, we look at oracle-free settings where $N$ decentralised agents learn along a single, non-episodic evolution path of the empirical system, such as we may encounter for a large range of many-agent cooperation problems in the real-world. We provide theoretical evidence that by spreading improved policies through the network in a decentralised fashion, our sample guarantees are upper-bounded by those of the purely independent-learning case. Moreover, we show empirically that our networked method can give faster convergence in practice, while removing the reliance on a centralised controller. We also demonstrate that our decentralised communication architecture brings significant benefits over both the centralised and independent alternatives in terms of robustness and flexibility to unexpected learning failures and changes in population size. For comparison purposes with our new architecture, we modify recent algorithms for the centralised and independent cases to make their practical convergence feasible: while contributing the first empirical demonstrations of these algorithms in our setting of $N$ agents learning along a single system evolution with only local state observability, we additionally display the empirical benefits of our new, networked approach.
☆ Tackling Non-Stationarity in Reinforcement Learning via Causal-Origin Representation
In real-world scenarios, the application of reinforcement learning is significantly challenged by complex non-stationarity. Most existing methods attempt to model the changes of the environment explicitly, often requiring impractical prior knowledge. In this paper, we propose a new perspective, positing that non-stationarity can propagate and accumulate through complex causal relationships during state transitions, thereby compounding its sophistication and affecting policy learning. We believe that this challenge can be more effectively addressed by tracing the causal origin of non-stationarity. To this end, we introduce the Causal-Origin REPresentation (COREP) algorithm. COREP primarily employs a guided updating mechanism to learn a stable graph representation for states termed as causal-origin representation. By leveraging this representation, the learned policy exhibits impressive resilience to non-stationarity. We supplement our approach with a theoretical analysis grounded in the causal interpretation for non-stationary reinforcement learning, advocating for the validity of the causal-origin representation. Experimental results further demonstrate the superior performance of COREP over existing methods in tackling non-stationarity.
☆ Towards Better Explanations for Object Detection
Recent advances in Artificial Intelligence (AI) technology have promoted their use in almost every field. The growing complexity of deep neural networks (DNNs) makes it increasingly difficult and important to explain the inner workings and decisions of the network. However, most current techniques for explaining DNNs focus mainly on interpreting classification tasks. This paper proposes a method to explain the decision for any object detection model called D-CLOSE. To closely track the model's behavior, we used multiple levels of segmentation on the image and a process to combine them. We performed tests on the MS-COCO dataset with the YOLOX model, which shows that our method outperforms D-RISE and can give a better quality and less noise explanation.
comment: 9 pages, 10 figures
☆ A Large-Scale Study of Probabilistic Calibration in Neural Network Regression
Accurate probabilistic predictions are essential for optimal decision making. While neural network miscalibration has been studied primarily in classification, we investigate this in the less-explored domain of regression. We conduct the largest empirical study to date to assess the probabilistic calibration of neural networks. We also analyze the performance of recalibration, conformal, and regularization methods to enhance probabilistic calibration. Additionally, we introduce novel differentiable recalibration and regularization methods, uncovering new insights into their effectiveness. Our findings reveal that regularization methods offer a favorable tradeoff between calibration and sharpness. Post-hoc methods exhibit superior probabilistic calibration, which we attribute to the finite-sample coverage guarantee of conformal prediction. Furthermore, we demonstrate that quantile recalibration can be considered as a specific case of conformal prediction. Our study is fully reproducible and implemented in a common code base for fair comparisons.
☆ Realising Synthetic Active Inference Agents, Part II: Variational Message Updates
The Free Energy Principle (FEP) describes (biological) agents as minimising a variational Free Energy (FE) with respect to a generative model of their environment. Active Inference (AIF) is a corollary of the FEP that describes how agents explore and exploit their environment by minimising an expected FE objective. In two related papers, we describe a scalable, epistemic approach to synthetic AIF agents, by message passing on free-form Forney-style Factor Graphs (FFGs). A companion paper (part I) introduces a Constrained FFG (CFFG) notation that visually represents (generalised) FE objectives for AIF. The current paper (part II) derives message passing algorithms that minimise (generalised) FE objectives on a CFFG by variational calculus. A comparison between simulated Bethe and generalised FE agents illustrates how synthetic AIF induces epistemic behaviour on a T-maze navigation task. With a full message passing account of synthetic AIF agents, it becomes possible to derive and reuse message updates across models and move closer to industrial applications of synthetic AIF.
☆ Conformal Prediction with Missing Values
Conformal prediction is a theoretically grounded framework for constructing predictive intervals. We study conformal prediction with missing values in the covariates -- a setting that brings new challenges to uncertainty quantification. We first show that the marginal coverage guarantee of conformal prediction holds on imputed data for any missingness distribution and almost all imputation functions. However, we emphasize that the average coverage varies depending on the pattern of missing values: conformal methods tend to construct prediction intervals that under-cover the response conditionally to some missing patterns. This motivates our novel generalized conformalized quantile regression framework, missing data augmentation, which yields prediction intervals that are valid conditionally to the patterns of missing values, despite their exponential number. We then show that a universally consistent quantile regression algorithm trained on the imputed data is Bayes optimal for the pinball risk, thus achieving valid coverage conditionally to any given data point. Moreover, we examine the case of a linear model, which demonstrates the importance of our proposal in overcoming the heteroskedasticity induced by missing values. Using synthetic and data from critical care, we corroborate our theory and report improved performance of our methods.
comment: Code for our experiments can be found at https://github.com/mzaffran/ConformalPredictionMissingValues . To be published in the proceedings of the 40th International Conference on Machine Learning, Honolulu, Hawaii, USA
☆ Enhanced Distribution Modelling via Augmented Architectures For Neural ODE Flows
While the neural ODE formulation of normalizing flows such as in FFJORD enables us to calculate the determinants of free form Jacobians in O(D) time, the flexibility of the transformation underlying neural ODEs has been shown to be suboptimal. In this paper, we present AFFJORD, a neural ODE-based normalizing flow which enhances the representation power of FFJORD by defining the neural ODE through special augmented transformation dynamics which preserve the topology of the space. Furthermore, we derive the Jacobian determinant of the general augmented form by generalizing the chain rule in the continuous sense into the Cable Rule, which expresses the forward sensitivity of ODEs with respect to their initial conditions. The cable rule gives an explicit expression for the Jacobian of a neural ODE transformation, and provides an elegant proof of the instantaneous change of variable. Our experimental results on density estimation in synthetic and high dimensional data, such as MNIST, CIFAR-10 and CelebA 32x32, show that AFFJORD outperforms the baseline FFJORD through the improved flexibility of the underlying vector field.
☆ Gibbs Sampling the Posterior of Neural Networks
In this paper, we study sampling from a posterior derived from a neural network. We propose a new probabilistic model consisting of adding noise at every pre- and post-activation in the network, arguing that the resulting posterior can be sampled using an efficient Gibbs sampler. The Gibbs sampler attains similar performances as the state-of-the-art Monte Carlo Markov chain methods, such as the Hamiltonian Monte Carlo or the Metropolis adjusted Langevin algorithm, both on real and synthetic data. By framing our analysis in the teacher-student setting, we introduce a thermalization criterion that allows us to detect when an algorithm, when run on data with synthetic labels, fails to sample from the posterior. The criterion is based on the fact that in the teacher-student setting we can initialize an algorithm directly at equilibrium.
☆ Multiple output samples for each input in a single-output Gaussian process
The standard Gaussian Process (GP) only considers a single output sample per input in the training set. Datasets for subjective tasks, such as spoken language assessment, may be annotated with output labels from multiple human raters per input. This paper proposes to generalise the GP to allow for these multiple output samples in the training set, and thus make use of available output uncertainty information. This differs from a multi-output GP, as all output samples are from the same task here. The output density function is formulated to be the joint likelihood of observing all output samples, and latent variables are not repeated to reduce computation cost. The test set predictions are inferred similarly to a standard GP, with a difference being in the optimised hyper-parameters. This is evaluated on speechocean762, showing that it allows the GP to compute a test set output distribution that is more similar to the collection of reference outputs from the multiple human raters.
☆ Federated Intrusion Detection System based on Deep Belief Networks
The vast increase of IoT technologies and the ever-evolving attack vectors and threat actors have increased cyber-security risks dramatically. Novel attacks can compromise IoT devices to gain access to sensitive data or control them to deploy further malicious activities. The detection of novel attacks often relies upon AI solutions. A common approach to implementing AI-based IDS in distributed IoT systems is in a centralised manner. However, this approach may violate data privacy and secrecy. In addition, centralised data collection prohibits the scale-up of IDSs. Therefore, intrusion detection solutions in IoT ecosystems need to move towards a decentralised direction. FL has attracted significant interest in recent years due to its ability to perform collaborative learning while preserving data confidentiality and locality. Nevertheless, most FL-based IDS for IoT systems are designed under unrealistic data distribution conditions. To that end, we design an experiment representative of the real world and evaluate the performance of two FL IDS implementations, one based on DNNs and another on our previous work on DBNs. For our experiments, we rely on TON-IoT, a realistic IoT network traffic dataset, associating each IP address with a single FL client. Additionally, we explore pre-training and investigate various aggregation methods to mitigate the impact of data heterogeneity. Lastly, we benchmark our approach against a centralised solution. The comparison shows that the heterogeneous nature of the data has a considerable negative impact on the model performance when trained in a distributed manner. However, in the case of a pre-trained initial global FL model, we demonstrate a performance improvement of over 20% (F1-score) when compared against a randomly initiated global model.
comment: 14 pages, 5 figues, 3 tables
☆ Comparative Study on Semi-supervised Learning Applied for Anomaly Detection in Hydraulic Condition Monitoring System
Condition-based maintenance is becoming increasingly important in hydraulic systems. However, anomaly detection for these systems remains challenging, especially since that anomalous data is scarce and labeling such data is tedious and even dangerous. Therefore, it is advisable to make use of unsupervised or semi-supervised methods, especially for semi-supervised learning which utilizes unsupervised learning as a feature extraction mechanism to aid the supervised part when only a small number of labels are available. This study systematically compares semi-supervised learning methods applied for anomaly detection in hydraulic condition monitoring systems. Firstly, thorough data analysis and feature learning were carried out to understand the open-sourced hydraulic condition monitoring dataset. Then, various methods were implemented and evaluated including traditional stand-alone semi-supervised learning models (e.g., one-class SVM, Robust Covariance), ensemble models (e.g., Isolation Forest), and deep neural network based models (e.g., autoencoder, Hierarchical Extreme Learning Machine (HELM)). Typically, this study customized and implemented an extreme learning machine based semi-supervised HELM model and verified its superiority over other semi-supervised methods. Extensive experiments show that the customized HELM model obtained state-of-the-art performance with the highest accuracy (99.5%), the lowest false positive rate (0.015), and the best F1-score (0.985) beating other semi-supervised methods.
comment: 7 pages, 8 figures, accepted by 2023 IEEE International Conference on Systems, Man, and Cybernetics (SMC 2023) https://ieeesmc2023.org/
☆ Orca: Progressive Learning from Complex Explanation Traces of GPT-4
Recent research has focused on enhancing the capability of smaller models through imitation learning, drawing on the outputs generated by large foundation models (LFMs). A number of issues impact the quality of these models, ranging from limited imitation signals from shallow LFM outputs; small scale homogeneous training data; and most notably a lack of rigorous evaluation resulting in overestimating the small model's capability as they tend to learn to imitate the style, but not the reasoning process of LFMs. To address these challenges, we develop Orca (We are working with our legal team to publicly release a diff of the model weights in accordance with LLaMA's release policy to be published at https://aka.ms/orca-lm), a 13-billion parameter model that learns to imitate the reasoning process of LFMs. Orca learns from rich signals from GPT-4 including explanation traces; step-by-step thought processes; and other complex instructions, guided by teacher assistance from ChatGPT. To promote this progressive learning, we tap into large-scale and diverse imitation data with judicious sampling and selection. Orca surpasses conventional state-of-the-art instruction-tuned models such as Vicuna-13B by more than 100% in complex zero-shot reasoning benchmarks like Big-Bench Hard (BBH) and 42% on AGIEval. Moreover, Orca reaches parity with ChatGPT on the BBH benchmark and shows competitive performance (4 pts gap with optimized system message) in professional and academic examinations like the SAT, LSAT, GRE, and GMAT, both in zero-shot settings without CoT; while trailing behind GPT-4. Our research indicates that learning from step-by-step explanations, whether these are generated by humans or more advanced AI models, is a promising direction to improve model capabilities and skills.
☆ Unlocking the Potential of Federated Learning for Deeper Models
Federated learning (FL) is a new paradigm for distributed machine learning that allows a global model to be trained across multiple clients without compromising their privacy. Although FL has demonstrated remarkable success in various scenarios, recent studies mainly utilize shallow and small neural networks. In our research, we discover a significant performance decline when applying the existing FL framework to deeper neural networks, even when client data are independently and identically distributed (i.i.d.). Our further investigation shows that the decline is due to the continuous accumulation of dissimilarities among client models during the layer-by-layer back-propagation process, which we refer to as "divergence accumulation." As deeper models involve a longer chain of divergence accumulation, they tend to manifest greater divergence, subsequently leading to performance decline. Both theoretical derivations and empirical evidence are proposed to support the existence of divergence accumulation and its amplified effects in deeper models. To address this issue, we propose several technical guidelines based on reducing divergence, such as using wider models and reducing the receptive field. These approaches can greatly improve the accuracy of FL on deeper models. For example, the application of these guidelines can boost the ResNet101 model's performance by as much as 43\% on the Tiny-ImageNet dataset.
comment: 16 pages, 8 figures
☆ Solving NP-hard Min-max Routing Problems as Sequential Generation with Equity Context
Min-max routing problems aim to minimize the maximum tour length among agents as they collaboratively visit all cities, i.e., the completion time. These problems include impactful real-world applications but are known as NP-hard. Existing methods are facing challenges, particularly in large-scale problems that require the coordination of numerous agents to cover thousands of cities. This paper proposes a new deep-learning framework to solve large-scale min-max routing problems. We model the simultaneous decision-making of multiple agents as a sequential generation process, allowing the utilization of scalable deep-learning models for sequential decision-making. In the sequentially approximated problem, we propose a scalable contextual Transformer model, Equity-Transformer, which generates sequential actions considering an equitable workload among other agents. The effectiveness of Equity-Transformer is demonstrated through its superior performance in two representative min-max routing tasks: the min-max multiple traveling salesman problem (min-max mTSP) and the min-max multiple pick-up and delivery problem (min-max mPDP). Notably, our method achieves significant reductions of runtime, approximately 335 times, and cost values of about 53% compared to a competitive heuristic (LKH3) in the case of 100 vehicles with 1,000 cities of mTSP. We provide reproducible source code: https://github.com/kaist-silab/equity-transformer
comment: 18 pages, 7 figures
☆ Meta-SAGE: Scale Meta-Learning Scheduled Adaptation with Guided Exploration for Mitigating Scale Shift on Combinatorial Optimization ICML
This paper proposes Meta-SAGE, a novel approach for improving the scalability of deep reinforcement learning models for combinatorial optimization (CO) tasks. Our method adapts pre-trained models to larger-scale problems in test time by suggesting two components: a scale meta-learner (SML) and scheduled adaptation with guided exploration (SAGE). First, SML transforms the context embedding for subsequent adaptation of SAGE based on scale information. Then, SAGE adjusts the model parameters dedicated to the context embedding for a specific instance. SAGE introduces locality bias, which encourages selecting nearby locations to determine the next location. The locality bias gradually decays as the model is adapted to the target instance. Results show that Meta-SAGE outperforms previous adaptation methods and significantly improves scalability in representative CO tasks. Our source code is available at https://github.com/kaist-silab/meta-sage
comment: 18 pages, 9 figures, International Conference on Machine Learning (ICML) 2023
☆ Predicting malaria dynamics in Burundi using deep Learning Models
Malaria continues to be a major public health problem on the African continent, particularly in Sub-Saharan Africa. Nonetheless, efforts are ongoing, and significant progress has been made. In Burundi, malaria is among the main public health concerns. In the literature, there are limited prediction models for Burundi. We know that such tools are much needed for interventions design. In our study, we built machine-learning based models to estimates malaria cases in Burundi. The forecast was carried out at province level, allowing us to estimate malaria cases on a national scale as well. Long short term memory (LSTM) model, a type of deep learning model has been used to achieve best results using climate-change related factors such as temperature, rainfal, and relative humidity, together with malaria historical data and human population. The results showed that at country level different tuning of parameters can be used in order to determine the minimum and maximum expected malaria
♻ ☆ STEVE-1: A Generative Model for Text-to-Behavior in Minecraft
Constructing AI models that respond to text instructions is challenging, especially for sequential decision-making tasks. This work introduces an instruction-tuned Video Pretraining (VPT) model for Minecraft called STEVE-1, demonstrating that the unCLIP approach, utilized in DALL-E 2, is also effective for creating instruction-following sequential decision-making agents. STEVE-1 is trained in two steps: adapting the pretrained VPT model to follow commands in MineCLIP's latent space, then training a prior to predict latent codes from text. This allows us to finetune VPT through self-supervised behavioral cloning and hindsight relabeling, bypassing the need for costly human text annotations. By leveraging pretrained models like VPT and MineCLIP and employing best practices from text-conditioned image generation, STEVE-1 costs just $60 to train and can follow a wide range of short-horizon open-ended text and visual instructions in Minecraft. STEVE-1 sets a new bar for open-ended instruction following in Minecraft with low-level controls (mouse and keyboard) and raw pixel inputs, far outperforming previous baselines. We provide experimental evidence highlighting key factors for downstream performance, including pretraining, classifier-free guidance, and data scaling. All resources, including our model weights, training scripts, and evaluation tools are made available for further research.
♻ ☆ FilFL: Client Filtering for Optimized Client Participation in Federated Learning
Federated learning is an emerging machine learning paradigm that enables clients to train collaboratively without exchanging local data. The clients participating in the training process have a crucial impact on the convergence rate, learning efficiency, and model generalization. In this work, we propose FilFL, a new approach to optimizing client participation and training by introducing client filtering. FilFL periodically filters the available clients to identify a subset that maximizes a combinatorial objective function using an efficient greedy filtering algorithm. From this filtered-in subset, clients are then selected for the training process. We provide a thorough analysis of FilFL convergence in a heterogeneous setting and evaluate its performance across diverse vision and language tasks and realistic federated scenarios with time-varying client availability. Our empirical results demonstrate several benefits of our approach, including improved learning efficiency, faster convergence, and up to 10 percentage points higher test accuracy compared to scenarios where client filtering is not utilized.
♻ ☆ In-context Example Selection with Influences
In-context learning (ICL) is a powerful paradigm emerged from large language models (LLMs). Despite its promises, ICL performance is known to be highly sensitive to input examples. In this work, we use $\textit{in-context influences}$ to analyze few-shot ICL performance directly from the in-context examples. Our proposed influence-based example selection method can identify both positive and negative examples, outperforming several baselines when evaluated on 9 SuperGLUE tasks. Our analysis uncovers up to a $16.3\%$ performance gap between using the most negative in-context examples compared to the most positive. In a case study, we apply our influence-based framework to quantify the phenomena of recency bias in example ordering for few-shot ICL.
♻ ☆ CLIP-Dissect: Automatic Description of Neuron Representations in Deep Vision Networks ICLR 2023
In this paper, we propose CLIP-Dissect, a new technique to automatically describe the function of individual hidden neurons inside vision networks. CLIP-Dissect leverages recent advances in multimodal vision/language models to label internal neurons with open-ended concepts without the need for any labeled data or human examples. We show that CLIP-Dissect provides more accurate descriptions than existing methods for last layer neurons where the ground-truth is available as well as qualitatively good descriptions for hidden layer neurons. In addition, our method is very flexible: it is model agnostic, can easily handle new concepts and can be extended to take advantage of better multimodal models in the future. Finally CLIP-Dissect is computationally efficient and can label all neurons from five layers of ResNet-50 in just 4 minutes, which is more than 10 times faster than existing methods. Our code is available at https://github.com/Trustworthy-ML-Lab/CLIP-dissect. Finally, crowdsourced user study results are available at Appendix B to further support the effectiveness of our method.
comment: Published in ICLR 2023 Conference (Spotlight). New v5(5 June 2023) - Added crowdsourced user study in Appendix B, not included in ICLR publication
♻ ☆ Centralized Training with Hybrid Execution in Multi-Agent Reinforcement Learning
We introduce hybrid execution in multi-agent reinforcement learning (MARL), a new paradigm in which agents aim to successfully complete cooperative tasks with arbitrary communication levels at execution time by taking advantage of information-sharing among the agents. Under hybrid execution, the communication level can range from a setting in which no communication is allowed between agents (fully decentralized), to a setting featuring full communication (fully centralized), but the agents do not know beforehand which communication level they will encounter at execution time. To formalize our setting, we define a new class of multi-agent partially observable Markov decision processes (POMDPs) that we name hybrid-POMDPs, which explicitly model a communication process between the agents. We contribute MARO, an approach that makes use of an auto-regressive predictive model, trained in a centralized manner, to estimate missing agents' observations at execution time. We evaluate MARO on standard scenarios and extensions of previous benchmarks tailored to emphasize the negative impact of partial observability in MARL. Experimental results show that our method consistently outperforms relevant baselines, allowing agents to act with faulty communication while successfully exploiting shared information.
♻ ☆ Label-Free Concept Bottleneck Models ICLR 2023
Concept bottleneck models (CBM) are a popular way of creating more interpretable neural networks by having hidden layer neurons correspond to human-understandable concepts. However, existing CBMs and their variants have two crucial limitations: first, they need to collect labeled data for each of the predefined concepts, which is time consuming and labor intensive; second, the accuracy of a CBM is often significantly lower than that of a standard neural network, especially on more complex datasets. This poor performance creates a barrier for adopting CBMs in practical real world applications. Motivated by these challenges, we propose Label-free CBM which is a novel framework to transform any neural network into an interpretable CBM without labeled concept data, while retaining a high accuracy. Our Label-free CBM has many advantages, it is: scalable - we present the first CBM scaled to ImageNet, efficient - creating a CBM takes only a few hours even for very large datasets, and automated - training it for a new dataset requires minimal human effort. Our code is available at https://github.com/Trustworthy-ML-Lab/Label-free-CBM. Finally, in Appendix B we conduct a large scale user evaluation of the interpretability of our method.
comment: Published at ICLR 2023. New v2(5 June 2023): added crowdsourced human study in Appendix B
♻ ☆ PFNs4BO: In-Context Learning for Bayesian Optimization ICML 2023
In this paper, we use Prior-data Fitted Networks (PFNs) as a flexible surrogate for Bayesian Optimization (BO). PFNs are neural processes that are trained to approximate the posterior predictive distribution (PPD) through in-context learning on any prior distribution that can be efficiently sampled from. We describe how this flexibility can be exploited for surrogate modeling in BO. We use PFNs to mimic a naive Gaussian process (GP), an advanced GP, and a Bayesian Neural Network (BNN). In addition, we show how to incorporate further information into the prior, such as allowing hints about the position of optima (user priors), ignoring irrelevant dimensions, and performing non-myopic BO by learning the acquisition function. The flexibility underlying these extensions opens up vast possibilities for using PFNs for BO. We demonstrate the usefulness of PFNs for BO in a large-scale evaluation on artificial GP samples and three different hyperparameter optimization testbeds: HPO-B, Bayesmark, and PD1. We publish code alongside trained models at https://github.com/automl/PFNs4BO.
comment: Accepted at ICML 2023
♻ ☆ Cooperative Open-ended Learning Framework for Zero-shot Coordination
Zero-shot coordination in cooperative artificial intelligence (AI) remains a significant challenge, which means effectively coordinating with a wide range of unseen partners. Previous algorithms have attempted to address this challenge by optimizing fixed objectives within a population to improve strategy or behaviour diversity. However, these approaches can result in a loss of learning and an inability to cooperate with certain strategies within the population, known as cooperative incompatibility. To address this issue, we propose the Cooperative Open-ended LEarning (COLE) framework, which constructs open-ended objectives in cooperative games with two players from the perspective of graph theory to assess and identify the cooperative ability of each strategy. We further specify the framework and propose a practical algorithm that leverages knowledge from game theory and graph theory. Furthermore, an analysis of the learning process of the algorithm shows that it can efficiently overcome cooperative incompatibility. The experimental results in the Overcooked game environment demonstrate that our method outperforms current state-of-the-art methods when coordinating with different-level partners. Our demo is available at https://sites.google.com/view/cole-2023.
comment: 15 pages with 9 pages main body
♻ ☆ Transformer and Snowball Graph Convolution Learning for Brain functional network Classification
Advanced deep learning methods, especially graph neural networks (GNNs), are increasingly expected to learn from brain functional network data and identify the functional connections between brain disorder and health. In this paper, we proposed a novel Transformer and snowball encoding networks (TSEN) for brain functional network classification, which introduced Transformer architecture with graph snowball connection into GNNs for learning whole-graph representation. TSEN combined graph snowball connection with graph Transformer by snowball encoding layers, which enhanced the power to capture multi-scale information and global patterns of brain functional networks. TSEN also introduced snowball graph convolution as position embedding in Transformer structure, which was a simple yet effective method for capturing local patterns naturally. We evaluated the proposed model by two large-scale brain functional network datasets, and the results demonstrated that TSEN outperformed the state-of-the-art GNN models and the graph-transformer based GNN models.
comment: Prepared for submitting to HBP
♻ ☆ Ambiguous Dynamic Treatment Regimes: A Reinforcement Learning Approach
A main research goal in various studies is to use an observational data set and provide a new set of counterfactual guidelines that can yield causal improvements. Dynamic Treatment Regimes (DTRs) are widely studied to formalize this process. However, available methods in finding optimal DTRs often rely on assumptions that are violated in real-world applications (e.g., medical decision-making or public policy), especially when (a) the existence of unobserved confounders cannot be ignored, and (b) the unobserved confounders are time-varying (e.g., affected by previous actions). When such assumptions are violated, one often faces ambiguity regarding the underlying causal model. This ambiguity is inevitable, since the dynamics of unobserved confounders and their causal impact on the observed part of the data cannot be understood from the observed data. Motivated by a case study of finding superior treatment regimes for patients who underwent transplantation in our partner hospital and faced a medical condition known as New Onset Diabetes After Transplantation (NODAT), we extend DTRs to a new class termed Ambiguous Dynamic Treatment Regimes (ADTRs), in which the causal impact of treatment regimes is evaluated based on a "cloud" of causal models. We then connect ADTRs to Ambiguous Partially Observable Mark Decision Processes (APOMDPs) and develop Reinforcement Learning methods, which enable using the observed data to efficiently learn an optimal treatment regime. We establish theoretical results for these learning methods, including (weak) consistency and asymptotic normality. We further evaluate the performance of these learning methods both in our case study and in simulation experiments.
♻ ☆ Action Noise in Off-Policy Deep Reinforcement Learning: Impact on Exploration and Performance
Many Deep Reinforcement Learning (D-RL) algorithms rely on simple forms of exploration such as the additive action noise often used in continuous control domains. Typically, the scaling factor of this action noise is chosen as a hyper-parameter and is kept constant during training. In this paper, we focus on action noise in off-policy deep reinforcement learning for continuous control. We analyze how the learned policy is impacted by the noise type, noise scale, and impact scaling factor reduction schedule. We consider the two most prominent types of action noise, Gaussian and Ornstein-Uhlenbeck noise, and perform a vast experimental campaign by systematically varying the noise type and scale parameter, and by measuring variables of interest like the expected return of the policy and the state-space coverage during exploration. For the latter, we propose a novel state-space coverage measure $\operatorname{X}_{\mathcal{U}\text{rel}}$ that is more robust to estimation artifacts caused by points close to the state-space boundary than previously-proposed measures. Larger noise scales generally increase state-space coverage. However, we found that increasing the space coverage using a larger noise scale is often not beneficial. On the contrary, reducing the noise scale over the training process reduces the variance and generally improves the learning performance. We conclude that the best noise type and scale are environment dependent, and based on our observations derive heuristic rules for guiding the choice of the action noise as a starting point for further optimization.
comment: Published in Transactions on Machine Learning Research (11/2022) https://openreview.net/forum?id=NljBlZ6hmG
♻ ☆ A Systematic Study and Comprehensive Evaluation of ChatGPT on Benchmark Datasets ACL 2023
The development of large language models (LLMs) such as ChatGPT has brought a lot of attention recently. However, their evaluation in the benchmark academic datasets remains under-explored due to the difficulty of evaluating the generative outputs produced by this model against the ground truth. In this paper, we aim to present a thorough evaluation of ChatGPT's performance on diverse academic datasets, covering tasks like question-answering, text summarization, code generation, commonsense reasoning, mathematical problem-solving, machine translation, bias detection, and ethical considerations. Specifically, we evaluate ChatGPT across 140 tasks and analyze 255K responses it generates in these datasets. This makes our work the largest evaluation of ChatGPT in NLP benchmarks. In short, our study aims to validate the strengths and weaknesses of ChatGPT in various tasks and provide insights for future research using LLMs. We also report a new emergent ability to follow multi-query instructions that we mostly found in ChatGPT and other instruction-tuned models. Our extensive evaluation shows that even though ChatGPT is capable of performing a wide variety of tasks, and may obtain impressive performance in several benchmark datasets, it is still far from achieving the ability to reliably solve many challenging tasks. By providing a thorough assessment of ChatGPT's performance across diverse NLP tasks, this paper sets the stage for a targeted deployment of ChatGPT-like LLMs in real-world applications.
comment: Accepted by ACL 2023 Findings. The first three authors contributed equally
♻ ☆ Reduce, Reuse, Recycle: Compositional Generation with Energy-Based Diffusion Models and MCMC ICML 2023
Since their introduction, diffusion models have quickly become the prevailing approach to generative modeling in many domains. They can be interpreted as learning the gradients of a time-varying sequence of log-probability density functions. This interpretation has motivated classifier-based and classifier-free guidance as methods for post-hoc control of diffusion models. In this work, we build upon these ideas using the score-based interpretation of diffusion models, and explore alternative ways to condition, modify, and reuse diffusion models for tasks involving compositional generation and guidance. In particular, we investigate why certain types of composition fail using current techniques and present a number of solutions. We conclude that the sampler (not the model) is responsible for this failure and propose new samplers, inspired by MCMC, which enable successful compositional generation. Further, we propose an energy-based parameterization of diffusion models which enables the use of new compositional operators and more sophisticated, Metropolis-corrected samplers. Intriguingly we find these samplers lead to notable improvements in compositional generation across a wide set of problems such as classifier-guided ImageNet modeling and compositional text-to-image generation.
comment: ICML 2023, Project Webpage: https://energy-based-model.github.io/reduce-reuse-recycle/
♻ ☆ XAudit : A Theoretical Look at Auditing with Explanations
Responsible use of machine learning requires models to be audited for undesirable properties. While a body of work has proposed using explanations for auditing, how to do so and why has remained relatively ill-understood. This work formalizes the role of explanations in auditing and investigates if and how model explanations can help audits. Specifically, we propose explanation-based algorithms for auditing linear classifiers and decision trees for feature sensitivity. Our results illustrate that Counterfactual explanations are extremely helpful for auditing. While Anchors and decision paths may not be as beneficial in the worst-case, in the average-case they do aid a lot.
♻ ☆ Robust incremental learning pipelines for temporal tabular datasets with distribution shifts
In this paper, we present a robust incremental learning model for regression tasks on temporal tabular datasets. Using commonly available tabular and time-series prediction models as building blocks, a machine-learning model is built incrementally to adapt to distributional shifts in data. Using the concept of self-similarity, the model uses only two basic building blocks of machine learning models, gradient boosting decision trees and neural networks to build models for any required complexity. The model is efficient as no specialised neural architectures are used and each model building block can be independently trained in parallel. The model is demonstrated to have robust performances under adverse situations such as regime changes, fat-tailed distributions and low signal-to-noise ratios. Model robustness are studied under different hyper-parameters and complexities.
♻ ☆ \textit{spred}: Solving $L_1$ Penalty with SGD ICML 2023
We propose to minimize a generic differentiable objective with $L_1$ constraint using a simple reparametrization and straightforward stochastic gradient descent. Our proposal is the direct generalization of previous ideas that the $L_1$ penalty may be equivalent to a differentiable reparametrization with weight decay. We prove that the proposed method, \textit{spred}, is an exact differentiable solver of $L_1$ and that the reparametrization trick is completely ``benign" for a generic nonconvex function. Practically, we demonstrate the usefulness of the method in (1) training sparse neural networks to perform gene selection tasks, which involves finding relevant features in a very high dimensional space, and (2) neural network compression task, to which previous attempts at applying the $L_1$-penalty have been unsuccessful. Conceptually, our result bridges the gap between the sparsity in deep learning and conventional statistical learning.
comment: ICML 2023, 16 pages, 10 figures, and 2 tables
♻ ☆ MiDi: Mixed Graph and 3D Denoising Diffusion for Molecule Generation
This work introduces MiDi, a novel diffusion model for jointly generating molecular graphs and their corresponding 3D arrangement of atoms. Unlike existing methods that rely on predefined rules to determine molecular bonds based on the 3D conformation, MiDi offers an end-to-end differentiable approach that streamlines the molecule generation process. Our experimental results demonstrate the effectiveness of this approach. On the challenging GEOM-DRUGS dataset, MiDi generates 92% of stable molecules, against 6% for the previous EDM model that uses interatomic distances for bond prediction, and 40% using EDM followed by an algorithm that directly optimize bond orders for validity. Our code is available at github.com/cvignac/MiDi.
comment: 22 pages. Under review
♻ ☆ Exploring the Connection between Robust and Generative Models
We offer a study that connects robust discriminative classifiers trained with adversarial training (AT) with generative modeling in the form of Energy-based Models (EBM). We do so by decomposing the loss of a discriminative classifier and showing that the discriminative model is also aware of the input data density. Though a common assumption is that adversarial points leave the manifold of the input data, our study finds out that, surprisingly, untargeted adversarial points in the input space are very likely under the generative model hidden inside the discriminative classifier -- have low energy in the EBM. We present two evidence: untargeted attacks are even more likely than the natural data and their likelihood increases as the attack strength increases. This allows us to easily detect them and craft a novel attack called High-Energy PGD that fools the classifier yet has energy similar to the data set. The code is available at github.com/senad96/Robust-Generative
comment: Italian Conference on AI - AI per Cybersecurity, 6 pages, 6 figures
♻ ☆ Adaptive Identification of Populations with Treatment Benefit in Clinical Trials: Machine Learning Challenges and Solutions
We study the problem of adaptively identifying patient subpopulations that benefit from a given treatment during a confirmatory clinical trial. This type of adaptive clinical trial has been thoroughly studied in biostatistics, but has been allowed only limited adaptivity so far. Here, we aim to relax classical restrictions on such designs and investigate how to incorporate ideas from the recent machine learning literature on adaptive and online experimentation to make trials more flexible and efficient. We find that the unique characteristics of the subpopulation selection problem -- most importantly that (i) one is usually interested in finding subpopulations with any treatment benefit (and not necessarily the single subgroup with largest effect) given a limited budget and that (ii) effectiveness only has to be demonstrated across the subpopulation on average -- give rise to interesting challenges and new desiderata when designing algorithmic solutions. Building on these findings, we propose AdaGGI and AdaGCPI, two meta-algorithms for subpopulation construction. We empirically investigate their performance across a range of simulation scenarios and derive insights into their (dis)advantages across different settings.
comment: To appear in the Proceedings of the 40th International Conference on Machine Learning, Honolulu, Hawaii, USA. PMLR 202, 2023
♻ ☆ Linking generative semi-supervised learning and generative open-set recognition
This study investigates the relationship between semi-supervised learning (SSL) and open-set recognition (OSR) in the context of generative adversarial networks (GANs). Although no previous study has formally linked SSL and OSR, their respective methods share striking similarities. Specifically, SSL-GANs and OSR-GANs require their generators to produce samples in the complementary space. Subsequently, by regularising networks with generated samples, both SSL and OSR classifiers generalize the open space. To demonstrate the connection between SSL and OSR, we theoretically and experimentally compare state-of-the-art SSL-GAN methods with state-of-the-art OSR-GAN methods. Our results indicate that the SSL optimised margin-GANs, which have a stronger foundation in literature, set the new standard for the combined SSL-OSR task and achieves new state-of-other art results in certain general OSR experiments. However, the OSR optimised adversarial reciprocal point (ARP)-GANs still slightly out-performed margin-GANs at other OSR experiments. This result indicates unique insights for the combined optimisation task of SSL-OSR.
♻ ☆ Deep Weakly-supervised Anomaly Detection KDD 2023
Recent semi-supervised anomaly detection methods that are trained using small labeled anomaly examples and large unlabeled data (mostly normal data) have shown largely improved performance over unsupervised methods. However, these methods often focus on fitting abnormalities illustrated by the given anomaly examples only (i.e.,, seen anomalies), and consequently they fail to generalize to those that are not, i.e., new types/classes of anomaly unseen during training. To detect both seen and unseen anomalies, we introduce a novel deep weakly-supervised approach, namely Pairwise Relation prediction Network (PReNet), that learns pairwise relation features and anomaly scores by predicting the relation of any two randomly sampled training instances, in which the pairwise relation can be anomaly-anomaly, anomaly-unlabeled, or unlabeled-unlabeled. Since unlabeled instances are mostly normal, the relation prediction enforces a joint learning of anomaly-anomaly, anomaly-normal, and normal-normal pairwise discriminative patterns, respectively. PReNet can then detect any seen/unseen abnormalities that fit the learned pairwise abnormal patterns, or deviate from the normal patterns. Further, this pairwise approach also seamlessly and significantly augments the training anomaly data. Empirical results on 12 real-world datasets show that PReNet significantly outperforms nine competing methods in detecting seen and unseen anomalies. We also theoretically and empirically justify the robustness of our model w.r.t. anomaly contamination in the unlabeled data. The code is available at https://github.com/mala-lab/PReNet.
comment: Accepted to KDD 2023
♻ ☆ Topology Optimization via Machine Learning and Deep Learning: A Review
Topology optimization (TO) is a method of deriving an optimal design that satisfies a given load and boundary conditions within a design domain. This method enables effective design without initial design, but has been limited in use due to high computational costs. At the same time, machine learning (ML) methodology including deep learning has made great progress in the 21st century, and accordingly, many studies have been conducted to enable effective and rapid optimization by applying ML to TO. Therefore, this study reviews and analyzes previous research on ML-based TO (MLTO). Two different perspectives of MLTO are used to review studies: (1) TO and (2) ML perspectives. The TO perspective addresses "why" to use ML for TO, while the ML perspective addresses "how" to apply ML to TO. In addition, the limitations of current MLTO research and future research directions are examined.
♻ ☆ Deep surrogate accelerated delayed-acceptance HMC for Bayesian inference of spatio-temporal heat fluxes in rotating disc systems
We introduce a deep learning accelerated methodology to solve PDE-based Bayesian inverse problems with guaranteed accuracy. This is motivated by the ill-posed problem of inferring a spatio-temporal heat-flux parameter known as the Biot number given temperature data, however the methodology is generalisable to other settings. To accelerate Bayesian inference, we develop a novel training scheme that uses data to adaptively train a neural-network surrogate simulating the parametric forward model. By simultaneously identifying an approximate posterior distribution over the Biot number, and weighting a physics-informed training loss according to this, our approach approximates forward and inverse solution together without any need for external solves. Using a random Chebyshev series, we outline how to approximate a Gaussian process prior, and using the surrogate we apply Hamiltonian Monte Carlo (HMC) to sample from the posterior distribution. We derive convergence of the surrogate posterior to the true posterior distribution in the Hellinger metric as our adaptive loss approaches zero. Additionally, we describe how this surrogate-accelerated HMC approach can be combined with traditional PDE solvers in a delayed-acceptance scheme to a-priori control the posterior accuracy. This overcomes a major limitation of deep learning-based surrogate approaches, which do not achieve guaranteed accuracy a-priori due to their non-convex training. Biot number calculations are involved in turbo-machinery design, which is safety critical and highly regulated, therefore it is important that our results have such mathematical guarantees. Our approach achieves fast mixing in high dimensions whilst retaining the convergence guarantees of a traditional PDE solver, and without the burden of evaluating this solver for proposals that are likely to be rejected. Numerical results are given using real and simulated data.
♻ ☆ Linkless Link Prediction via Relational Distillation
Graph Neural Networks (GNNs) have shown exceptional performance in the task of link prediction. Despite their effectiveness, the high latency brought by non-trivial neighborhood data dependency limits GNNs in practical deployments. Conversely, the known efficient MLPs are much less effective than GNNs due to the lack of relational knowledge. In this work, to combine the advantages of GNNs and MLPs, we start with exploring direct knowledge distillation (KD) methods for link prediction, i.e., predicted logit-based matching and node representation-based matching. Upon observing direct KD analogs do not perform well for link prediction, we propose a relational KD framework, Linkless Link Prediction (LLP), to distill knowledge for link prediction with MLPs. Unlike simple KD methods that match independent link logits or node representations, LLP distills relational knowledge that is centered around each (anchor) node to the student MLP. Specifically, we propose rank-based matching and distribution-based matching strategies that complement each other. Extensive experiments demonstrate that LLP boosts the link prediction performance of MLPs with significant margins, and even outperforms the teacher GNNs on 7 out of 8 benchmarks. LLP also achieves a 70.68x speedup in link prediction inference compared to GNNs on the large-scale OGB dataset.
♻ ☆ MAMO: Masked Multimodal Modeling for Fine-Grained Vision-Language Representation Learning SIGIR 2023
Multimodal representation learning has shown promising improvements on various vision-language tasks. Most existing methods excel at building global-level alignment between vision and language while lacking effective fine-grained image-text interaction. In this paper, we propose a jointly masked multimodal modeling method to learn fine-grained multimodal representations. Our method performs joint masking on image-text input and integrates both implicit and explicit targets for the masked signals to recover. The implicit target provides a unified and debiased objective for vision and language, where the model predicts latent multimodal representations of the unmasked input. The explicit target further enriches the multimodal representations by recovering high-level and semantically meaningful information: momentum visual features of image patches and concepts of word tokens. Through such a masked modeling process, our model not only learns fine-grained multimodal interaction, but also avoids the semantic gap between high-level representations and low- or mid-level prediction targets (e.g. image pixels), thus producing semantically rich multimodal representations that perform well on both zero-shot and fine-tuned settings. Our pre-trained model (named MAMO) achieves state-of-the-art performance on various downstream vision-language tasks, including image-text retrieval, visual question answering, visual reasoning, and weakly-supervised visual grounding.
comment: SIGIR 2023, 10 pages
♻ ☆ Robust and Generalisable Segmentation of Subtle Epilepsy-causing Lesions: a Graph Convolutional Approach MICCAI 2023
Focal cortical dysplasia (FCD) is a leading cause of drug-resistant focal epilepsy, which can be cured by surgery. These lesions are extremely subtle and often missed even by expert neuroradiologists. "Ground truth" manual lesion masks are therefore expensive, limited and have large inter-rater variability. Existing FCD detection methods are limited by high numbers of false positive predictions, primarily due to vertex- or patch-based approaches that lack whole-brain context. Here, we propose to approach the problem as semantic segmentation using graph convolutional networks (GCN), which allows our model to learn spatial relationships between brain regions. To address the specific challenges of FCD identification, our proposed model includes an auxiliary loss to predict distance from the lesion to reduce false positives and a weak supervision classification loss to facilitate learning from uncertain lesion masks. On a multi-centre dataset of 1015 participants with surface-based features and manual lesion masks from structural MRI data, the proposed GCN achieved an AUC of 0.74, a significant improvement against a previously used vertex-wise multi-layer perceptron (MLP) classifier (AUC 0.64). With sensitivity thresholded at 67%, the GCN had a specificity of 71% in comparison to 49% when using the MLP. This improvement in specificity is vital for clinical integration of lesion-detection tools into the radiological workflow, through increasing clinical confidence in the use of AI radiological adjuncts and reducing the number of areas requiring expert review.
comment: accepted at MICCAI 2023
♻ ☆ A Robust Optimisation Perspective on Counterexample-Guided Repair of Neural Networks ICML 2023
Counterexample-guided repair aims at creating neural networks with mathematical safety guarantees, facilitating the application of neural networks in safety-critical domains. However, whether counterexample-guided repair is guaranteed to terminate remains an open question. We approach this question by showing that counterexample-guided repair can be viewed as a robust optimisation algorithm. While termination guarantees for neural network repair itself remain beyond our reach, we prove termination for more restrained machine learning models and disprove termination in a general setting. We empirically study the practical implications of our theoretical results, demonstrating the suitability of common verifiers and falsifiers for repair despite a disadvantageous theoretical result. Additionally, we use our theoretical insights to devise a novel algorithm for repairing linear regression models based on quadratic programming, surpassing existing approaches.
comment: Accepted at ICML 2023. 9 pages + 13 pages appendix, 8 figures
♻ ☆ Sequoia: A Software Framework to Unify Continual Learning Research
The field of Continual Learning (CL) seeks to develop algorithms that accumulate knowledge and skills over time through interaction with non-stationary environments. In practice, a plethora of evaluation procedures (settings) and algorithmic solutions (methods) exist, each with their own potentially disjoint set of assumptions. This variety makes measuring progress in CL difficult. We propose a taxonomy of settings, where each setting is described as a set of assumptions. A tree-shaped hierarchy emerges from this view, where more general settings become the parents of those with more restrictive assumptions. This makes it possible to use inheritance to share and reuse research, as developing a method for a given setting also makes it directly applicable onto any of its children. We instantiate this idea as a publicly available software framework called Sequoia, which features a wide variety of settings from both the Continual Supervised Learning (CSL) and Continual Reinforcement Learning (CRL) domains. Sequoia also includes a growing suite of methods which are easy to extend and customize, in addition to more specialized methods from external libraries. We hope that this new paradigm and its first implementation can help unify and accelerate research in CL. You can help us grow the tree by visiting www.github.com/lebrice/Sequoia.
♻ ☆ Deep Reinforcement Learning with Swin Transformers
Transformers are neural network models that utilize multiple layers of self-attention heads and have exhibited enormous potential in natural language processing tasks. Meanwhile, there have been efforts to adapt transformers to visual tasks of machine learning, including Vision Transformers and Swin Transformers. Although some researchers use Vision Transformers for reinforcement learning tasks, their experiments remain at a small scale due to the high computational cost. Experiments conducted at a large scale, on the other hand, have to rely on techniques to cut the costs of Vision Transformers, which also yield inferior results. To address this challenge, this article presents the first online reinforcement learning scheme that is based on Swin Transformers: Swin DQN. Swin Transformers are promising as a backbone in neural networks by splitting groups of image pixels into small patches and applying local self-attention operations inside the (shifted) windows of fixed sizes. They have demonstrated state-of-the-art performances in benchmarks. In contrast to existing research, our novel approach is reducing the computational costs, as well as significantly improving the performance. We demonstrate the superior performance with experiments on 49 games in the Arcade Learning Environment. The results show that our approach, using Swin Transformers with Double DQN, achieves significantly higher maximal evaluation scores than the baseline method in 45 of all the 49 games ~92%, and higher mean evaluation scores than the baseline method in 40 of all the 49 games ~82%.
♻ ☆ Learning Prototype Classifiers for Long-Tailed Recognition IJCAI-23
The problem of long-tailed recognition (LTR) has received attention in recent years due to the fundamental power-law distribution of objects in the real-world. Most recent works in LTR use softmax classifiers that have a tendency to correlate classifier norm with the amount of training data for a given class. On the other hand, Prototype classifiers do not suffer from this shortcoming and can deliver promising results simply using Nearest-Class-Mean (NCM), a special case where prototypes are empirical centroids. However, the potential of Prototype classifiers as an alternative to softmax in LTR is relatively underexplored. In this work, we propose Prototype classifiers, which jointly learn prototypes that minimize average cross-entropy loss based on probability scores from distances to prototypes. We theoretically analyze the properties of Euclidean distance based prototype classifiers that leads to stable gradient-based optimization which is robust to outliers. We further enhance Prototype classifiers by learning channel-dependent temperature parameters to enable independent distance scales along each channel. Our analysis shows that prototypes learned by Prototype classifiers are better separated than empirical centroids. Results on four long-tailed recognition benchmarks show that Prototype classifier outperforms or is comparable to the state-of-the-art methods.
comment: Accepted at IJCAI-23
♻ ☆ Tensor Completion with Provable Consistency and Fairness Guarantees for Recommender Systems
We introduce a new consistency-based approach for defining and solving nonnegative/positive matrix and tensor completion problems. The novelty of the framework is that instead of artificially making the problem well-posed in the form of an application-arbitrary optimization problem, e.g., minimizing a bulk structural measure such as rank or norm, we show that a single property/constraint: preserving unit-scale consistency, guarantees the existence of both a solution and, under relatively weak support assumptions, uniqueness. The framework and solution algorithms also generalize directly to tensors of arbitrary dimensions while maintaining computational complexity that is linear in problem size for fixed dimension d. In the context of recommender system (RS) applications, we prove that two reasonable properties that should be expected to hold for any solution to the RS problem are sufficient to permit uniqueness guarantees to be established within our framework. Key theoretical contributions include a general unit-consistent tensor-completion framework with proofs of its properties, e.g., consensus-order and fairness, and algorithms with optimal runtime and space complexities, e.g., O(1) term-completion with preprocessing complexity that is linear in the number of known terms of the matrix/tensor. From a practical perspective, the seamless ability of the framework to generalize to exploit high-dimensional structural relationships among key state variables, e.g., user and product attributes, offers a means for extracting significantly more information than is possible for alternative methods that cannot generalize beyond direct user-product relationships. Finally, we propose our consensus ordering property as an admissibility criterion for any proposed RS method.
comment: Final revision after acceptance by journal
♻ ☆ RLSbench: Domain Adaptation Under Relaxed Label Shift ICML 2023
Despite the emergence of principled methods for domain adaptation under label shift, their sensitivity to shifts in class conditional distributions is precariously under explored. Meanwhile, popular deep domain adaptation heuristics tend to falter when faced with label proportions shifts. While several papers modify these heuristics in attempts to handle label proportions shifts, inconsistencies in evaluation standards, datasets, and baselines make it difficult to gauge the current best practices. In this paper, we introduce RLSbench, a large-scale benchmark for relaxed label shift, consisting of $>$500 distribution shift pairs spanning vision, tabular, and language modalities, with varying label proportions. Unlike existing benchmarks, which primarily focus on shifts in class-conditional $p(x|y)$, our benchmark also focuses on label marginal shifts. First, we assess 13 popular domain adaptation methods, demonstrating more widespread failures under label proportion shifts than were previously known. Next, we develop an effective two-step meta-algorithm that is compatible with most domain adaptation heuristics: (i) pseudo-balance the data at each epoch; and (ii) adjust the final classifier with target label distribution estimate. The meta-algorithm improves existing domain adaptation heuristics under large label proportion shifts, often by 2--10\% accuracy points, while conferring minimal effect ($<$0.5\%) when label proportions do not shift. We hope that these findings and the availability of RLSbench will encourage researchers to rigorously evaluate proposed methods in relaxed label shift settings. Code is publicly available at https://github.com/acmi-lab/RLSbench.
comment: Accepted at ICML 2023. Paper website: https://sites.google.com/view/rlsbench/
♻ ☆ Data Poisoning Attacks Against Multimodal Encoders
Recently, the newly emerged multimodal models, which leverage both visual and linguistic modalities to train powerful encoders, have gained increasing attention. However, learning from a large-scale unlabeled dataset also exposes the model to the risk of potential poisoning attacks, whereby the adversary aims to perturb the model's training data to trigger malicious behaviors in it. In contrast to previous work, only poisoning visual modality, in this work, we take the first step to studying poisoning attacks against multimodal models in both visual and linguistic modalities. Specially, we focus on answering two questions: (1) Is the linguistic modality also vulnerable to poisoning attacks? and (2) Which modality is most vulnerable? To answer the two questions, we propose three types of poisoning attacks against multimodal models. Extensive evaluations on different datasets and model architectures show that all three attacks can achieve significant attack performance while maintaining model utility in both visual and linguistic modalities. Furthermore, we observe that the poisoning effect differs between different modalities. To mitigate the attacks, we propose both pre-training and post-training defenses. We empirically show that both defenses can significantly reduce the attack performance while preserving the model's utility.
comment: To Appear in the 40th International Conference on Machine Learning, July 2023
♻ ☆ Differentially Private Episodic Reinforcement Learning with Heavy-tailed Rewards ICML 2023
In this paper, we study the problem of (finite horizon tabular) Markov decision processes (MDPs) with heavy-tailed rewards under the constraint of differential privacy (DP). Compared with the previous studies for private reinforcement learning that typically assume rewards are sampled from some bounded or sub-Gaussian distributions to ensure DP, we consider the setting where reward distributions have only finite $(1+v)$-th moments with some $v \in (0,1]$. By resorting to robust mean estimators for rewards, we first propose two frameworks for heavy-tailed MDPs, i.e., one is for value iteration and another is for policy optimization. Under each framework, we consider both joint differential privacy (JDP) and local differential privacy (LDP) models. Based on our frameworks, we provide regret upper bounds for both JDP and LDP cases and show that the moment of distribution and privacy budget both have significant impacts on regrets. Finally, we establish a lower bound of regret minimization for heavy-tailed MDPs in JDP model by reducing it to the instance-independent lower bound of heavy-tailed multi-armed bandits in DP model. We also show the lower bound for the problem in LDP by adopting some private minimax methods. Our results reveal that there are fundamental differences between the problem of private RL with sub-Gaussian and that with heavy-tailed rewards.
comment: ICML 2023
♻ ☆ Tactile-Filter: Interactive Tactile Perception for Part Mating
Humans rely on touch and tactile sensing for a lot of dexterous manipulation tasks. Our tactile sensing provides us with a lot of information regarding contact formations as well as geometric information about objects during any interaction. With this motivation, vision-based tactile sensors are being widely used for various robotic perception and control tasks. In this paper, we present a method for interactive perception using vision-based tactile sensors for a part mating task, where a robot can use tactile sensors and a feedback mechanism using a particle filter to incrementally improve its estimate of objects (pegs and holes) that fit together. To do this, we first train a deep neural network that makes use of tactile images to predict the probabilistic correspondence between arbitrarily shaped objects that fit together. The trained model is used to design a particle filter which is used twofold. First, given one partial (or non-unique) observation of the hole, it incrementally improves the estimate of the correct peg by sampling more tactile observations. Second, it selects the next action for the robot to sample the next touch (and thus image) which results in maximum uncertainty reduction to minimize the number of interactions during the perception task. We evaluate our method on several part-mating tasks with novel objects using a robot equipped with a vision-based tactile sensor. We also show the efficiency of the proposed action selection method against a naive method. See supplementary video at https://www.youtube.com/watch?v=jMVBg_e3gLw .
comment: Accepted at RSS2023
♻ ☆ Mitigating Off-Policy Bias in Actor-Critic Methods with One-Step Q-learning: A Novel Correction Approach
Compared to on-policy counterparts, off-policy model-free deep reinforcement learning can improve data efficiency by repeatedly using the previously gathered data. However, off-policy learning becomes challenging when the discrepancy between the underlying distributions of the agent's policy and collected data increases. Although the well-studied importance sampling and off-policy policy gradient techniques were proposed to compensate for this discrepancy, they usually require a collection of long trajectories and induce additional problems such as vanishing/exploding gradients or discarding many useful experiences, which eventually increases the computational complexity. Moreover, their generalization to either continuous action domains or policies approximated by deterministic deep neural networks is strictly limited. To overcome these limitations, we introduce a novel policy similarity measure to mitigate the effects of such discrepancy in continuous control. Our method offers an adequate single-step off-policy correction that is applicable to deterministic policy networks. Theoretical and empirical studies demonstrate that it can achieve a "safe" off-policy learning and substantially improve the state-of-the-art by attaining higher returns in fewer steps than the competing methods through an effective schedule of the learning rate in Q-learning and policy optimization.
♻ ☆ Lightweight and Flexible Deep Equilibrium Learning for CSI Feedback in FDD Massive MIMO
In frequency-division duplexing (FDD) massive multiple-input multiple-output (MIMO) systems, downlink channel state information (CSI) needs to be sent back to the base station (BS) by the users, which causes prohibitive feedback overhead. In this paper, we propose a lightweight and flexible deep learning-based CSI feedback approach by capitalizing on deep equilibrium models. Different from existing deep learning-based methods that stack multiple explicit layers, we propose an implicit equilibrium block to mimic the behavior of an infinite-depth neural network. In particular, the implicit equilibrium block is defined by a fixed-point iteration and the trainable parameters in different iterations are shared, which results in a lightweight model. Furthermore, the number of forward iterations can be adjusted according to users' computation capability, enabling a flexible accuracy-efficiency trade-off. Simulation results will show that the proposed design obtains a comparable performance as the benchmarks but with much-reduced complexity and permits an accuracy-efficiency trade-off at runtime.
comment: submitted to IEEE for possible publication
♻ ☆ OpenFE: Automated Feature Generation with Expert-level Performance ICML2023
The goal of automated feature generation is to liberate machine learning experts from the laborious task of manual feature generation, which is crucial for improving the learning performance of tabular data. The major challenge in automated feature generation is to efficiently and accurately identify effective features from a vast pool of candidate features. In this paper, we present OpenFE, an automated feature generation tool that provides competitive results against machine learning experts. OpenFE achieves high efficiency and accuracy with two components: 1) a novel feature boosting method for accurately evaluating the incremental performance of candidate features and 2) a two-stage pruning algorithm that performs feature pruning in a coarse-to-fine manner. Extensive experiments on ten benchmark datasets show that OpenFE outperforms existing baseline methods by a large margin. We further evaluate OpenFE in two Kaggle competitions with thousands of data science teams participating. In the two competitions, features generated by OpenFE with a simple baseline model can beat 99.3% and 99.6% data science teams respectively. In addition to the empirical results, we provide a theoretical perspective to show that feature generation can be beneficial in a simple yet representative setting. The code is available at https://github.com/ZhangTP1996/OpenFE.
comment: 22 pages, 3 figures, accepted by ICML2023
♻ ☆ Exploiting Noise as a Resource for Computation and Learning in Spiking Neural Networks
Networks of spiking neurons underpin the extraordinary information-processing capabilities of the brain and have emerged as pillar models in neuromorphic intelligence. Despite extensive research on spiking neural networks (SNNs), most are established on deterministic models. Integrating noise into SNNs leads to biophysically more realistic neural dynamics and may benefit model performance. This work presents the noisy spiking neural network (NSNN) and the noise-driven learning rule (NDL) by introducing a spiking neuron model incorporating noisy neuronal dynamics. Our approach shows how noise may act as a resource for computation and learning and theoretically provides a framework for general SNNs. Moreover, NDL provides an insightful biological rationale for surrogate gradients. By incorporating various SNN architectures and algorithms, we show that our approach exhibits competitive performance and improved robustness against challenging perturbations than deterministic SNNs. Additionally, we demonstrate the utility of the NSNN model for neural coding studies. Overall, NSNN offers a powerful, flexible, and easy-to-use tool for machine learning practitioners and computational neuroscience researchers.
comment: Updated the code link; fixed the bug in the BBL file generated with bibliography management program
♻ ☆ Existence and Estimation of Critical Batch Size for Training Generative Adversarial Networks with Two Time-Scale Update Rule ICML 2023
Previous results have shown that a two time-scale update rule (TTUR) using different learning rates, such as different constant rates or different decaying rates, is useful for training generative adversarial networks (GANs) in theory and in practice. Moreover, not only the learning rate but also the batch size is important for training GANs with TTURs and they both affect the number of steps needed for training. This paper studies the relationship between batch size and the number of steps needed for training GANs with TTURs based on constant learning rates. We theoretically show that, for a TTUR with constant learning rates, the number of steps needed to find stationary points of the loss functions of both the discriminator and generator decreases as the batch size increases and that there exists a critical batch size minimizing the stochastic first-order oracle (SFO) complexity. Then, we use the Fr'echet inception distance (FID) as the performance measure for training and provide numerical results indicating that the number of steps needed to achieve a low FID score decreases as the batch size increases and that the SFO complexity increases once the batch size exceeds the measured critical batch size. Moreover, we show that measured critical batch sizes are close to the sizes estimated from our theoretical results.
comment: Accepted at the 40th International Conference on Machine Learning (ICML 2023)
♻ ☆ Canonical foliations of neural networks: application to robustness
Deep learning models are known to be vulnerable to adversarial attacks. Adversarial learning is therefore becoming a crucial task. We propose a new vision on neural network robustness using Riemannian geometry and foliation theory. The idea is illustrated by creating a new adversarial attack that takes into account the curvature of the data space. This new adversarial attack called the two-step spectral attack is a piece-wise linear approximation of a geodesic in the data space. The data space is treated as a (degenerate) Riemannian manifold equipped with the pullback of the Fisher Information Metric (FIM) of the neural network. In most cases, this metric is only semi-definite and its kernel becomes a central object to study. A canonical foliation is derived from this kernel. The curvature of transverse leaves gives the appropriate correction to get a two-step approximation of the geodesic and hence a new efficient adversarial attack. The method is first illustrated on a 2D toy example in order to visualize the neural network foliation and the corresponding attacks. Next, experiments on the MNIST dataset with the proposed technique and a state of the art attack presented in Zhao et al. (2019) are reported. The result show that the proposed attack is more efficient at all levels of available budget for the attack (norm of the attack), confirming that the curvature of the transverse neural network FIM foliation plays an important role in the robustness of neural networks.
♻ ☆ Scalable Optimal Margin Distribution Machine
Optimal margin Distribution Machine (ODM) is a newly proposed statistical learning framework rooting in the novel margin theory, which demonstrates better generalization performance than the traditional large margin based counterparts. Nonetheless, it suffers from the ubiquitous scalability problem regarding both computation time and memory as other kernel methods. This paper proposes a scalable ODM, which can achieve nearly ten times speedup compared to the original ODM training method. For nonlinear kernels, we propose a novel distribution-aware partition method to make the local ODM trained on each partition be close and converge fast to the global one. When linear kernel is applied, we extend a communication efficient SVRG method to accelerate the training further. Extensive empirical studies validate that our proposed method is highly computational efficient and almost never worsen the generalization.
♻ ☆ Evaluation of Interpretability Methods and Perturbation Artifacts in Deep Neural Networks
Despite excellent performance of deep neural networks (DNNs) in image classification, detection, and prediction, characterizing how DNNs make a given decision remains an open problem, resulting in a number of interpretability methods. Post-hoc interpretability methods primarily aim to quantify the importance of input features with respect to the class probabilities. However, due to the lack of ground truth and the existence of interpretability methods with diverse operating characteristics, evaluating these methods is a crucial challenge. A popular approach to evaluate interpretability methods is to perturb input features deemed important for a given prediction and observe the decrease in accuracy. However, perturbation itself may introduce artifacts. We propose a method for estimating the impact of such artifacts on the fidelity estimation by utilizing model accuracy curves from perturbing input features according to the Most Import First (MIF) and Least Import First (LIF) orders. Using the ResNet-50 trained on the ImageNet, we demonstrate the proposed fidelity estimation of four popular post-hoc interpretability methods.
comment: 11 pages, 5 figures
♻ ☆ Exponential Smoothing for Off-Policy Learning ICML 2023
Off-policy learning (OPL) aims at finding improved policies from logged bandit data, often by minimizing the inverse propensity scoring (IPS) estimator of the risk. In this work, we investigate a smooth regularization for IPS, for which we derive a two-sided PAC-Bayes generalization bound. The bound is tractable, scalable, interpretable and provides learning certificates. In particular, it is also valid for standard IPS without making the assumption that the importance weights are bounded. We demonstrate the relevance of our approach and its favorable performance through a set of learning tasks. Since our bound holds for standard IPS, we are able to provide insight into when regularizing IPS is useful. Namely, we identify cases where regularization might not be needed. This goes against the belief that, in practice, clipped IPS often enjoys favorable performance than standard IPS in OPL.
comment: ICML 2023 (Oral and Poster)
♻ ☆ Signs of Language: Embodied Sign Language Fingerspelling Acquisition from Demonstrations for Human-Robot Interaction
Learning fine-grained movements is a challenging topic in robotics, particularly in the context of robotic hands. One specific instance of this challenge is the acquisition of fingerspelling sign language in robots. In this paper, we propose an approach for learning dexterous motor imitation from video examples without additional information. To achieve this, we first build a URDF model of a robotic hand with a single actuator for each joint. We then leverage pre-trained deep vision models to extract the 3D pose of the hand from RGB videos. Next, using state-of-the-art reinforcement learning algorithms for motion imitation (namely, proximal policy optimization and soft actor-critic), we train a policy to reproduce the movement extracted from the demonstrations. We identify the optimal set of hyperparameters for imitation based on a reference motion. Finally, we demonstrate the generalizability of our approach by testing it on six different tasks, corresponding to fingerspelled letters. Our results show that our approach is able to successfully imitate these fine-grained movements without additional information, highlighting its potential for real-world applications in robotics.
♻ ☆ Scaling Multi-Objective Security Games Provably via Space Discretization Based Evolutionary Search
In the field of security, multi-objective security games (MOSGs) allow defenders to simultaneously protect targets from multiple heterogeneous attackers. MOSGs aim to simultaneously maximize all the heterogeneous payoffs, e.g., life, money, and crime rate, without merging heterogeneous attackers. In real-world scenarios, the number of heterogeneous attackers and targets to be protected may exceed the capability of most existing state-of-the-art methods, i.e., MOSGs are limited by the issue of scalability. To this end, this paper proposes a general framework called SDES based on many-objective evolutionary search to scale up MOSGs to large-scale targets and heterogeneous attackers. SDES consists of four consecutive key components, i.e., discretization, optimization, evaluation, and refinement. Specifically, SDES first discretizes the originally high-dimensional continuous solution space to the low-dimensional discrete one by the maximal indifference property in game theory. This property helps evolutionary algorithms (EAs) bypass the high-dimensional step function and ensure a well-convergent Pareto front. Then, a many-objective EA is used for optimization in the low-dimensional discrete solution space to obtain a well-spaced Pareto front. To evaluate solutions, SDES restores solutions back to the original space via greedily optimizing a novel divergence measurement. Finally, the refinement in SDES boosts the optimization performance with acceptable cost. Theoretically, we prove the optimization consistency and convergence of SDES. Experiment results show that SDES is the first linear-time MOSG algorithm for both large-scale attackers and targets. SDES is able to solve up to 20 attackers and 100 targets MOSG problems, while the state-of-the-art (SOTA) methods can only solve up to 8 attackers and 25 targets ones. Ablation study verifies the necessity of all components in SDES.
♻ ☆ Construction of Hierarchical Neural Architecture Search Spaces based on Context-free Grammars
The discovery of neural architectures from simple building blocks is a long-standing goal of Neural Architecture Search (NAS). Hierarchical search spaces are a promising step towards this goal but lack a unifying search space design framework and typically only search over some limited aspect of architectures. In this work, we introduce a unifying search space design framework based on context-free grammars that can naturally and compactly generate expressive hierarchical search spaces that are 100s of orders of magnitude larger than common spaces from the literature. By enhancing and using their properties, we effectively enable search over the complete architecture and can foster regularity. Further, we propose an efficient hierarchical kernel design for a Bayesian Optimization search strategy to efficiently search over such huge spaces. We demonstrate the versatility of our search space design framework and show that our search strategy can be superior to existing NAS approaches. Code is available at https://github.com/automl/hierarchical_nas_construction.
♻ ☆ Graph Neural Networks are Inherently Good Generalizers: Insights by Bridging GNNs and MLPs ICLR 2023
Graph neural networks (GNNs), as the de-facto model class for representation learning on graphs, are built upon the multi-layer perceptrons (MLP) architecture with additional message passing layers to allow features to flow across nodes. While conventional wisdom commonly attributes the success of GNNs to their advanced expressivity, we conjecture that this is not the main cause of GNNs' superiority in node-level prediction tasks. This paper pinpoints the major source of GNNs' performance gain to their intrinsic generalization capability, by introducing an intermediate model class dubbed as P(ropagational)MLP, which is identical to standard MLP in training, but then adopts GNN's architecture in testing. Intriguingly, we observe that PMLPs consistently perform on par with (or even exceed) their GNN counterparts, while being much more efficient in training. This finding sheds new insights into understanding the learning behavior of GNNs, and can be used as an analytic tool for dissecting various GNN-related research problems. As an initial step to analyze the inherent generalizability of GNNs, we show the essential difference between MLP and PMLP at infinite-width limit lies in the NTK feature map in the post-training stage. Moreover, by examining their extrapolation behavior, we find that though many GNNs and their PMLP counterparts cannot extrapolate non-linear functions for extremely out-of-distribution samples, they have greater potential to generalize to testing samples near the training data range as natural advantages of GNN architectures.
comment: Accepted to ICLR 2023. Codes in https://github.com/chr26195/PMLP
♻ ☆ Safe Peeling for L0-Regularized Least-Squares with supplementary material
We introduce a new methodology dubbed ``safe peeling'' to accelerate the resolution of L0-regularized least-squares problems via a Branch-and-Bound (BnB) algorithm. Our procedure enables to tighten the convex relaxation considered at each node of the BnB decision tree and therefore potentially allows for more aggressive pruning. Numerical simulations show that our proposed methodology leads to significant gains in terms of number of nodes explored and overall solving time.s show that our proposed methodology leads to significant gains in terms of number of nodes explored and overall solving time.
♻ ☆ Learning the Relation between Similarity Loss and Clustering Loss in Self-Supervised Learning
Self-supervised learning enables networks to learn discriminative features from massive data itself. Most state-of-the-art methods maximize the similarity between two augmentations of one image based on contrastive learning. By utilizing the consistency of two augmentations, the burden of manual annotations can be freed. Contrastive learning exploits instance-level information to learn robust features. However, the learned information is probably confined to different views of the same instance. In this paper, we attempt to leverage the similarity between two distinct images to boost representation in self-supervised learning. In contrast to instance-level information, the similarity between two distinct images may provide more useful information. Besides, we analyze the relation between similarity loss and feature-level cross-entropy loss. These two losses are essential for most deep learning methods. However, the relation between these two losses is not clear. Similarity loss helps obtain instance-level representation, while feature-level cross-entropy loss helps mine the similarity between two distinct images. We provide theoretical analyses and experiments to show that a suitable combination of these two losses can get state-of-the-art results. Code is available at https://github.com/guijiejie/ICCL.
comment: This paper is accepted by IEEE Transactions on Image Processing
♻ ☆ Learning Failure-Inducing Models for Testing Software-Defined Networks
Software-defined networks (SDN) enable flexible and effective communication systems that are managed by centralized software controllers. However, such a controller can undermine the underlying communication network of an SDN-based system and thus must be carefully tested. When an SDN-based system fails, in order to address such a failure, engineers need to precisely understand the conditions under which it occurs. In this article, we introduce a machine learning-guided fuzzing method, named FuzzSDN, aiming at both (1) generating effective test data leading to failures in SDN-based systems and (2) learning accurate failure-inducing models that characterize conditions under which such system fails. To our knowledge, FuzzSDN is the first attempt to simultaneously address these two objectives for SDNs. We evaluate FuzzSDN by applying it to systems controlled by two open-source SDN controllers. Further, we compare FuzzSDN with two state-of-the-art methods for fuzzing SDNs and two baselines for learning failure-inducing models. Our results show that (1) compared to the state-of-the-art methods, FuzzSDN generates at least 12 times more failures, within the same time budget, with a controller that is fairly robust to fuzzing and (2) our failure-inducing models have, on average, a precision of 98% and a recall of 86%, significantly outperforming the baselines.
♻ ☆ An adaptive safety layer with hard constraints for safe reinforcement learning in multi-energy management systems
Safe reinforcement learning (RL) with hard constraint guarantees is a promising optimal control direction for multi-energy management systems. It only requires the environment-specific constraint functions itself a priori and not a complete model (i.e. plant, disturbance and noise models, and prediction models for states not included in the plant model - e.g. demand forecasts, weather forecasts, price forecasts). The project-specific upfront and ongoing engineering efforts are therefore still reduced, better representations of the underlying system dynamics can still be learned and modelling bias is kept to a minimum (no model-based objective function). However, even the constraint functions alone are not always trivial to accurately provide in advance, leading to potentially unsafe behaviour. In this paper, we present two novel advancements: (I) combining the Optlayer and SafeFallback method, named OptLayerPolicy, to increase the initial utility while keeping a high sample efficiency. (II) introducing self-improving hard constraints, to increase the accuracy of the constraint functions as more data becomes available so that better policies can be learned. Both advancements keep the constraint formulation decoupled from the RL formulation, so that new (presumably better) RL algorithms can act as drop-in replacements. We have shown that, in a simulated multi-energy system case study, the initial utility is increased to 92.4% (OptLayerPolicy) compared to 86.1% (OptLayer) and that the policy after training is increased to 104.9% (GreyOptLayerPolicy) compared to 103.4% (OptLayer) - all relative to a vanilla RL benchmark. While introducing surrogate functions into the optimization problem requires special attention, we do conclude that the newly presented GreyOptLayerPolicy method is the most advantageous.
comment: 4703 words. arXiv admin note: text overlap with arXiv:2207.03830
♻ ☆ ACTC: Active Threshold Calibration for Cold-Start Knowledge Graph Completion ACL'23
Self-supervised knowledge-graph completion (KGC) relies on estimating a scoring model over (entity, relation, entity)-tuples, for example, by embedding an initial knowledge graph. Prediction quality can be improved by calibrating the scoring model, typically by adjusting the prediction thresholds using manually annotated examples. In this paper, we attempt for the first time cold-start calibration for KGC, where no annotated examples exist initially for calibration, and only a limited number of tuples can be selected for annotation. Our new method ACTC finds good per-relation thresholds efficiently based on a limited set of annotated tuples. Additionally to a few annotated tuples, ACTC also leverages unlabeled tuples by estimating their correctness with Logistic Regression or Gaussian Process classifiers. We also experiment with different methods for selecting candidate tuples for annotation: density-based and random selection. Experiments with five scoring models and an oracle annotator show an improvement of 7% points when using ACTC in the challenging setting with an annotation budget of only 10 tuples, and an average improvement of 4% points over different budgets.
comment: Accepted to ACL'23
♻ ☆ Pre-training for Speech Translation: CTC Meets Optimal Transport ICML 2023
The gap between speech and text modalities is a major challenge in speech-to-text translation (ST). Different methods have been proposed to reduce this gap, but most of them require architectural changes in ST training. In this work, we propose to mitigate this issue at the pre-training stage, requiring no change in the ST model. First, we show that the connectionist temporal classification (CTC) loss can reduce the modality gap by design. We provide a quantitative comparison with the more common cross-entropy loss, showing that pre-training with CTC consistently achieves better final ST accuracy. Nevertheless, CTC is only a partial solution and thus, in our second contribution, we propose a novel pre-training method combining CTC and optimal transport to further reduce this gap. Our method pre-trains a Siamese-like model composed of two encoders, one for acoustic inputs and the other for textual inputs, such that they produce representations that are close to each other in the Wasserstein space. Extensive experiments on the standard CoVoST-2 and MuST-C datasets show that our pre-training method applied to the vanilla encoder-decoder Transformer achieves state-of-the-art performance under the no-external-data setting, and performs on par with recent strong multi-task learning systems trained with external data. Finally, our method can also be applied on top of these multi-task systems, leading to further improvements for these models. Code and pre-trained models are available at https://github.com/formiel/fairseq.
comment: ICML 2023 (oral presentation). This version fixed URLs, updated affiliations & acknowledgements, and improved formatting
♻ ☆ Candidate Set Re-ranking for Composed Image Retrieval with Dual Multi-modal Encoder
Composed image retrieval aims to find an image that best matches a given multi-modal user query consisting of a reference image and text pair. Existing methods commonly pre-compute image embeddings over the entire corpus and compare these to a reference image embedding modified by the query text at test time. Such a pipeline is very efficient at test time since fast vector distances can be used to evaluate candidates, but modifying the reference image embedding guided only by a short textual description can be difficult, especially independent of potential candidates. An alternative approach is to allow interactions between the query and every possible candidate, i.e., reference-text-candidate triplets, and pick the best from the entire set. Though this approach is more discriminative, for large-scale datasets the computational cost is prohibitive since pre-computation of candidate embeddings is no longer possible. We propose to combine the merits of both schemes using a two-stage model. Our first stage adopts the conventional vector distancing metric and performs a fast pruning among candidates. Meanwhile, our second stage employs a dual-encoder architecture, which effectively attends to the input triplet of reference-text-candidate and re-ranks the candidates. Both stages utilize a vision-and-language pre-trained network, which has proven beneficial for various downstream tasks. Our method consistently outperforms state-of-the-art approaches on standard benchmarks for the task.
comment: 14 pages
♻ ☆ Position tracking of a varying number of sound sources with sliding permutation invariant training
Recent data- and learning-based sound source localization (SSL) methods have shown strong performance in challenging acoustic scenarios. However, little work has been done on adapting such methods to track consistently multiple sources appearing and disappearing, as would occur in reality. In this paper, we present a new training strategy for deep learning SSL models with a straightforward implementation based on the mean squared error of the optimal association between estimated and reference positions in the preceding time frames. It optimizes the desired properties of a tracking system: handling a time-varying number of sources and ordering localization estimates according to their trajectories, minimizing identity switches (IDSs). Evaluation on simulated data of multiple reverberant moving sources and on two model architectures proves its effectiveness on reducing identity switches without compromising frame-wise localization accuracy.
comment: Accepted for publication at the 31st European Signal Processing Conference (EUSIPCO 2023)
♻ ☆ Centralised rehearsal of decentralised cooperation: Multi-agent reinforcement learning for the scalable coordination of residential energy flexibility
This paper investigates how deep multi-agent reinforcement learning can enable the scalable and privacy-preserving coordination of residential energy flexibility. The coordination of distributed resources such as electric vehicles and heating will be critical to the successful integration of large shares of renewable energy in our electricity grid and, thus, to help mitigate climate change. The pre-learning of individual reinforcement learning policies can enable distributed control with no sharing of personal data required during execution. However, previous approaches for multi-agent reinforcement learning-based distributed energy resources coordination impose an ever greater training computational burden as the size of the system increases. We therefore adopt a deep multi-agent actor-critic method which uses a \emph{centralised but factored critic} to rehearse coordination ahead of execution. Results show that coordination is achieved at scale, with minimal information and communication infrastructure requirements, no interference with daily activities, and privacy protection. Significant savings are obtained for energy users, the distribution network and greenhouse gas emissions. Moreover, training times are nearly 40 times shorter than with a previous state-of-the-art reinforcement learning approach without the factored critic for 30 homes.
♻ ☆ Evaluating Robustness and Uncertainty of Graph Models Under Structural Distributional Shifts
In reliable decision-making systems based on machine learning, models have to be robust to distributional shifts or provide the uncertainty of their predictions. In node-level problems of graph learning, distributional shifts can be especially complex since the samples are interdependent. To evaluate the performance of graph models, it is important to test them on diverse and meaningful distributional shifts. However, most graph benchmarks considering distributional shifts for node-level problems focus mainly on node features, while structural properties are also essential for graph problems. In this work, we propose a general approach for inducing diverse distributional shifts based on graph structure. We use this approach to create data splits according to several structural node properties: popularity, locality, and density. In our experiments, we thoroughly evaluate the proposed distributional shifts and show that they can be quite challenging for existing graph models. We also reveal that simple models often outperform more sophisticated methods on these challenging shifts. Finally, our experiments provide evidence that there is a trade-off between the quality of learned representations for the base classification task under structural distributional shift and the ability to separate the nodes from different distributions using these representations.
♻ ☆ Vandermonde Neural Operators
Fourier Neural Operators (FNOs) have emerged as very popular machine learning architectures for learning operators, particularly those arising in PDEs. However, as FNOs rely on the fast Fourier transform for computational efficiency, the architecture can be limited to input data on equispaced Cartesian grids. Here, we generalize FNOs to handle input data on non-equispaced point distributions. Our proposed model, termed as Vandermonde Neural Operator (VNO), utilizes Vandermonde-structured matrices to efficiently compute forward and inverse Fourier transforms, even on arbitrarily distributed points. We present numerical experiments to demonstrate that VNOs can be significantly faster than FNOs, while retaining comparable accuracy, and improve upon accuracy of comparable non-equispaced methods such as the Geo-FNO.
comment: 21 pages, 10 figures
♻ ☆ Rethinking Gauss-Newton for learning over-parameterized models
This work studies the global convergence and generalization properties of Gauss Newton's (GN) when optimizing one-hidden layer networks in the over-parameterized regime. We first establish a global convergence result for GN in the continuous-time limit exhibiting a faster convergence rate compared to GD due to improved conditioning. We then perform an empirical study on a synthetic regression task to investigate the implicit bias of GN's method. We find that, while GN is consistently faster than GD in finding a global optimum, the performance of the learned model on a test dataset is heavily influenced by both the learning rate and the variance of the randomly initialized network's weights. Specifically, we find that initializing with a smaller variance results in a better generalization, a behavior also observed for GD. However, in contrast to GD where larger learning rates lead to the best generalization, we find that GN achieves an improved generalization when using smaller learning rates, albeit at the cost of slower convergence. This study emphasizes the significance of the learning rate in balancing the optimization speed of GN with the generalization ability of the learned solution.
♻ ☆ Learning Activation Functions for Sparse Neural Networks
Sparse Neural Networks (SNNs) can potentially demonstrate similar performance to their dense counterparts while saving significant energy and memory at inference. However, the accuracy drop incurred by SNNs, especially at high pruning ratios, can be an issue in critical deployment conditions. While recent works mitigate this issue through sophisticated pruning techniques, we shift our focus to an overlooked factor: hyperparameters and activation functions. Our analyses have shown that the accuracy drop can additionally be attributed to (i) Using ReLU as the default choice for activation functions unanimously, and (ii) Fine-tuning SNNs with the same hyperparameters as dense counterparts. Thus, we focus on learning a novel way to tune activation functions for sparse networks and combining these with a separate hyperparameter optimization (HPO) regime for sparse networks. By conducting experiments on popular DNN models (LeNet-5, VGG-16, ResNet-18, and EfficientNet-B0) trained on MNIST, CIFAR-10, and ImageNet-16 datasets, we show that the novel combination of these two approaches, dubbed Sparse Activation Function Search, short: SAFS, results in up to 15.53%, 8.88%, and 6.33% absolute improvement in the accuracy for LeNet-5, VGG-16, and ResNet-18 over the default training protocols, especially at high pruning ratios. Our code can be found at https://github.com/automl/SAFS
♻ ☆ Causal Counterfactuals for Improving the Robustness of Reinforcement Learning AAMAS 2023
Reinforcement learning (RL) is used in various robotic applications. RL enables agents to learn tasks autonomously by interacting with the environment. The more critical the tasks are, the higher the demand for the robustness of the RL systems. Causal RL combines RL and causal inference to make RL more robust. Causal RL agents use a causal representation to capture the invariant causal mechanisms that can be transferred from one task to another. Currently, there is limited research in Causal RL, and existing solutions are usually not complete or feasible for real-world applications. In this work, we propose CausalCF, the first complete Causal RL solution incorporating ideas from Causal Curiosity and CoPhy. Causal Curiosity provides an approach for using interventions, and CoPhy is modified to enable the RL agent to perform counterfactuals. Causal Curiosity has been applied to robotic grasping and manipulation tasks in CausalWorld. CausalWorld provides a realistic simulation environment based on the TriFinger robot. We apply CausalCF to complex robotic tasks and show that it improves the RL agent's robustness using CausalWorld.
comment: Accepted to ARMS-2023 (ARMS-2023: AAMAS 2023 Workshop on Autonomous Robots and Multirobot Systems)
♻ ☆ Learning Meta Representations for Agents in Multi-Agent Reinforcement Learning
In multi-agent reinforcement learning, the behaviors that agents learn in a single Markov Game (MG) are typically confined to the given agent number. Every single MG induced by varying the population may possess distinct optimal joint strategies and game-specific knowledge, which are modeled independently in modern multi-agent reinforcement learning algorithms. In this work, our focus is on creating agents that can generalize across population-varying MGs. Instead of learning a unimodal policy, each agent learns a policy set comprising effective strategies across a variety of games. To achieve this, we propose Meta Representations for Agents (MRA) that explicitly models the game-common and game-specific strategic knowledge. By representing the policy sets with multi-modal latent policies, the game-common strategic knowledge and diverse strategic modes are discovered through an iterative optimization procedure. We prove that by approximately maximizing the resulting constrained mutual information objective, the policies can reach Nash Equilibrium in every evaluation MG when the latent space is sufficiently large. When deploying MRA in practical settings with limited latent space sizes, fast adaptation can be achieved by leveraging the first-order gradient information. Extensive experiments demonstrate the effectiveness of MRA in improving training performance and generalization ability in challenging evaluation games.
comment: Published at CoLLAs
♻ ☆ Prediction intervals for neural network models using weighted asymmetric loss functions
We propose a simple and efficient approach to generate prediction intervals (PIs) for approximated and forecasted trends. Our method leverages a weighted asymmetric loss function to estimate the lower and upper bounds of the PIs, with the weights determined by the interval width. We provide a concise mathematical proof of the method, show how it can be extended to derive PIs for parametrised functions and argue why the method works for predicting PIs of dependent variables. The presented tests of the method on a real-world forecasting task using a neural network-based model show that it can produce reliable PIs in complex machine learning scenarios.
comment: 14 pages, 3 figures, not submitted anywhere yet
♻ ☆ Discover and Cure: Concept-aware Mitigation of Spurious Correlation ICML 2023
Deep neural networks often rely on spurious correlations to make predictions, which hinders generalization beyond training environments. For instance, models that associate cats with bed backgrounds can fail to predict the existence of cats in other environments without beds. Mitigating spurious correlations is crucial in building trustworthy models. However, the existing works lack transparency to offer insights into the mitigation process. In this work, we propose an interpretable framework, Discover and Cure (DISC), to tackle the issue. With human-interpretable concepts, DISC iteratively 1) discovers unstable concepts across different environments as spurious attributes, then 2) intervenes on the training data using the discovered concepts to reduce spurious correlation. Across systematic experiments, DISC provides superior generalization ability and interpretability than the existing approaches. Specifically, it outperforms the state-of-the-art methods on an object recognition task and a skin-lesion classification task by 7.5% and 9.6%, respectively. Additionally, we offer theoretical analysis and guarantees to understand the benefits of models trained by DISC. Code and data are available at https://github.com/Wuyxin/DISC.
comment: ICML 2023
♻ ☆ Evaluating Generative Patent Language Models
Generative language models are promising for assisting human writing in various domains. This manuscript aims to build generative language models in the patent domain and evaluate model performance from a human-centric perspective. The perspective is to measure the ratio of keystrokes that can be saved by autocompletion based on generative patent language models. A higher ratio means a more effective model which can save more keystrokes. This metric can be used to benchmark model performance. The metric is different from conventional machine-centric metrics that are token-based instead of keystroke-based. In terms of model size, the largest model built in this manuscript is 6B, which is state-of-the-art in the patent domain. Based on the metric, it is found that the largest model is not necessarily the best for the human-centric metric. The finding means that keeping increasing model sizes in the patent domain might be unnecessary if the purpose is to assist human writing with autocompletion. Several patent language models are pre-trained from scratch in this research. The pre-trained models are released for future researchers. Several visualization tools are also provided. The importance of building a generative language model in the patent domain is the potential to facilitate creativity and innovations in the future.
comment: 12 pages, 7 figures, and 5 tables
♻ ☆ ClusterFuG: Clustering Fully connected Graphs by Multicut ICML 2023
We propose a graph clustering formulation based on multicut (a.k.a. weighted correlation clustering) on the complete graph. Our formulation does not need specification of the graph topology as in the original sparse formulation of multicut, making our approach simpler and potentially better performing. In contrast to unweighted correlation clustering we allow for a more expressive weighted cost structure. In dense multicut, the clustering objective is given in a factorized form as inner products of node feature vectors. This allows for an efficient formulation and inference in contrast to multicut/weighted correlation clustering, which has at least quadratic representation and computation complexity when working on the complete graph. We show how to rewrite classical greedy algorithms for multicut in our dense setting and how to modify them for greater efficiency and solution quality. In particular, our algorithms scale to graphs with tens of thousands of nodes. Empirical evidence on instance segmentation on Cityscapes and clustering of ImageNet datasets shows the merits of our approach.
comment: ICML 2023
♻ ☆ Revisiting Personalized Federated Learning: Robustness Against Backdoor Attacks KDD 2023
In this work, besides improving prediction accuracy, we study whether personalization could bring robustness benefits to backdoor attacks. We conduct the first study of backdoor attacks in the pFL framework, testing 4 widely used backdoor attacks against 6 pFL methods on benchmark datasets FEMNIST and CIFAR-10, a total of 600 experiments. The study shows that pFL methods with partial model-sharing can significantly boost robustness against backdoor attacks. In contrast, pFL methods with full model-sharing do not show robustness. To analyze the reasons for varying robustness performances, we provide comprehensive ablation studies on different pFL methods. Based on our findings, we further propose a lightweight defense method, Simple-Tuning, which empirically improves defense performance against backdoor attacks. We believe that our work could provide both guidance for pFL application in terms of its robustness and offer valuable insights to design more robust FL methods in the future. We open-source our code to establish the first benchmark for black-box backdoor attacks in pFL: https://github.com/alibaba/FederatedScope/tree/backdoor-bench.
comment: KDD 2023
♻ ☆ Beyond the Meta: Leveraging Game Design Parameters for Patch-Agnostic Esport Analytics
Esport games comprise a sizeable fraction of the global games market, and is the fastest growing segment in games. This has given rise to the domain of esports analytics, which uses telemetry data from games to inform players, coaches, broadcasters and other stakeholders. Compared to traditional sports, esport titles change rapidly, in terms of mechanics as well as rules. Due to these frequent changes to the parameters of the game, esport analytics models can have a short life-spam, a problem which is largely ignored within the literature. This paper extracts information from game design (i.e. patch notes) and utilises clustering techniques to propose a new form of character representation. As a case study, a neural network model is trained to predict the number of kills in a Dota 2 match utilising this novel character representation technique. The performance of this model is then evaluated against two distinct baselines, including conventional techniques. Not only did the model significantly outperform the baselines in terms of accuracy (85% AUC), but the model also maintains the accuracy in two newer iterations of the game that introduced one new character and a brand new character type. These changes introduced to the design of the game would typically break conventional techniques that are commonly used within the literature. Therefore, the proposed methodology for representing characters can increase the life-spam of machine learning models as well as contribute to a higher performance when compared to traditional techniques typically employed within the literature.
Multimedia 4
☆ Towards Unified Text-based Person Retrieval: A Large-scale Multi-Attribute and Language Search Benchmark
In this paper, we introduce a large Multi-Attribute and Language Search dataset for text-based person retrieval, called MALS, and explore the feasibility of performing pre-training on both attribute recognition and image-text matching tasks in one stone. In particular, MALS contains 1,510,330 image-text pairs, which is about 37.5 times larger than prevailing CUHK-PEDES, and all images are annotated with 27 attributes. Considering the privacy concerns and annotation costs, we leverage the off-the-shelf diffusion models to generate the dataset. To verify the feasibility of learning from the generated data, we develop a new joint Attribute Prompt Learning and Text Matching Learning (APTM) framework, considering the shared knowledge between attribute and text. As the name implies, APTM contains an attribute prompt learning stream and a text matching learning stream. (1) The attribute prompt learning leverages the attribute prompts for image-attribute alignment, which enhances the text matching learning. (2) The text matching learning facilitates the representation learning on fine-grained details, and in turn, boosts the attribute prompt learning. Extensive experiments validate the effectiveness of the pre-training on MALS, achieving state-of-the-art retrieval performance via APTM on three challenging real-world benchmarks. In particular, APTM achieves a consistent improvement of +6.60%, +7.39%, and +15.90% Recall@1 accuracy on CUHK-PEDES, ICFG-PEDES, and RSTPReid datasets by a clear margin, respectively.
☆ Do-GOOD: Towards Distribution Shift Evaluation for Pre-Trained Visual Document Understanding Models SIGIR 2023
Numerous pre-training techniques for visual document understanding (VDU) have recently shown substantial improvements in performance across a wide range of document tasks. However, these pre-trained VDU models cannot guarantee continued success when the distribution of test data differs from the distribution of training data. In this paper, to investigate how robust existing pre-trained VDU models are to various distribution shifts, we first develop an out-of-distribution (OOD) benchmark termed Do-GOOD for the fine-Grained analysis on Document image-related tasks specifically. The Do-GOOD benchmark defines the underlying mechanisms that result in different distribution shifts and contains 9 OOD datasets covering 3 VDU related tasks, e.g., document information extraction, classification and question answering. We then evaluate the robustness and perform a fine-grained analysis of 5 latest VDU pre-trained models and 2 typical OOD generalization algorithms on these OOD datasets. Results from the experiments demonstrate that there is a significant performance gap between the in-distribution (ID) and OOD settings for document images, and that fine-grained analysis of distribution shifts can reveal the brittle nature of existing pre-trained VDU models and OOD generalization algorithms. The code and datasets for our Do-GOOD benchmark can be found at https://github.com/MAEHCM/Do-GOOD.
comment: SIGIR 2023. The code and datasets for our Do-GOOD benchmark can be found at https://github.com/MAEHCM/Do-GOOD
♻ ☆ MAMO: Masked Multimodal Modeling for Fine-Grained Vision-Language Representation Learning SIGIR 2023
Multimodal representation learning has shown promising improvements on various vision-language tasks. Most existing methods excel at building global-level alignment between vision and language while lacking effective fine-grained image-text interaction. In this paper, we propose a jointly masked multimodal modeling method to learn fine-grained multimodal representations. Our method performs joint masking on image-text input and integrates both implicit and explicit targets for the masked signals to recover. The implicit target provides a unified and debiased objective for vision and language, where the model predicts latent multimodal representations of the unmasked input. The explicit target further enriches the multimodal representations by recovering high-level and semantically meaningful information: momentum visual features of image patches and concepts of word tokens. Through such a masked modeling process, our model not only learns fine-grained multimodal interaction, but also avoids the semantic gap between high-level representations and low- or mid-level prediction targets (e.g. image pixels), thus producing semantically rich multimodal representations that perform well on both zero-shot and fine-tuned settings. Our pre-trained model (named MAMO) achieves state-of-the-art performance on various downstream vision-language tasks, including image-text retrieval, visual question answering, visual reasoning, and weakly-supervised visual grounding.
comment: SIGIR 2023, 10 pages
♻ ☆ QuantArt: Quantizing Image Style Transfer Towards High Visual Fidelity CVPR 2023
The mechanism of existing style transfer algorithms is by minimizing a hybrid loss function to push the generated image toward high similarities in both content and style. However, this type of approach cannot guarantee visual fidelity, i.e., the generated artworks should be indistinguishable from real ones. In this paper, we devise a new style transfer framework called QuantArt for high visual-fidelity stylization. QuantArt pushes the latent representation of the generated artwork toward the centroids of the real artwork distribution with vector quantization. By fusing the quantized and continuous latent representations, QuantArt allows flexible control over the generated artworks in terms of content preservation, style similarity, and visual fidelity. Experiments on various style transfer settings show that our QuantArt framework achieves significantly higher visual fidelity compared with the existing style transfer methods.
comment: Accepted to CVPR 2023. Code is available at https://github.com/siyuhuang/QuantArt
Computation and Language 47
☆ RadLing: Towards Efficient Radiology Report Understanding ACL
Most natural language tasks in the radiology domain use language models pre-trained on biomedical corpus. There are few pretrained language models trained specifically for radiology, and fewer still that have been trained in a low data setting and gone on to produce comparable results in fine-tuning tasks. We present RadLing, a continuously pretrained language model using Electra-small (Clark et al., 2020) architecture, trained using over 500K radiology reports, that can compete with state-of-the-art results for fine tuning tasks in radiology domain. Our main contribution in this paper is knowledge-aware masking which is a taxonomic knowledge-assisted pretraining task that dynamically masks tokens to inject knowledge during pretraining. In addition, we also introduce an knowledge base-aided vocabulary extension to adapt the general tokenization vocabulary to radiology domain.
comment: Association for Computational Linguistics (ACL), 2023
☆ Modeling Cross-Cultural Pragmatic Inference with Codenames Duet ACL 2023
Pragmatic reference enables efficient interpersonal communication. Prior work uses simple reference games to test models of pragmatic reasoning, often with unidentified speakers and listeners. In practice, however, speakers' sociocultural background shapes their pragmatic assumptions. For example, readers of this paper assume NLP refers to "Natural Language Processing," and not "Neuro-linguistic Programming." This work introduces the Cultural Codes dataset, which operationalizes sociocultural pragmatic inference in a simple word reference game. Cultural Codes is based on the multi-turn collaborative two-player game, Codenames Duet. Our dataset consists of 794 games with 7,703 turns, distributed across 153 unique players. Alongside gameplay, we collect information about players' personalities, values, and demographics. Utilizing theories of communication and pragmatics, we predict each player's actions via joint modeling of their sociocultural priors and the game context. Our experiments show that accounting for background characteristics significantly improves model performance for tasks related to both clue giving and guessing, indicating that sociocultural priors play a vital role in gameplay decisions.
comment: ACL 2023 Findings
☆ Adaptive and Personalized Exercise Generation for Online Language Learning ACL 2023
Adaptive learning aims to provide customized educational activities (e.g., exercises) to address individual learning needs. However, manual construction and delivery of such activities is a laborious process. Thus, in this paper, we study a novel task of adaptive and personalized exercise generation for online language learning. To this end, we combine a knowledge tracing model that estimates each student's evolving knowledge states from their learning history and a controlled text generation model that generates exercise sentences based on the student's current estimated knowledge state and instructor requirements of desired properties (e.g., domain knowledge and difficulty). We train and evaluate our model on real-world learner interaction data from Duolingo and demonstrate that LMs guided by student states can generate superior exercises. Then, we discuss the potential use of our model in educational applications using various simulations. These simulations show that our model can adapt to students' individual abilities and can facilitate their learning efficiency by personalizing learning sequences.
comment: To appear at ACL 2023
☆ Taught by the Internet, Exploring Bias in OpenAIs GPT3
This research delves into the current literature on bias in Natural Language Processing Models and the techniques proposed to mitigate the problem of bias, including why it is important to tackle bias in the first place. Additionally, these techniques are further analysed in the light of newly developed models that tower in size over past editions. To achieve those aims, the authors of this paper conducted their research on GPT3 by OpenAI, the largest NLP model available to consumers today. With 175 billion parameters in contrast to BERTs 340 million, GPT3 is the perfect model to test the common pitfalls of NLP models. Tests were conducted through the development of an Applicant Tracking System using GPT3. For the sake of feasibility and time constraints, the tests primarily focused on gender bias, rather than all or multiple types of bias. Finally, current mitigation techniques are considered and tested to measure their degree of functionality.
☆ Evaluating and Improving Tool-Augmented Computation-Intensive Math Reasoning
Chain-of-thought prompting~(CoT) and tool augmentation have been validated in recent work as effective practices for improving large language models~(LLMs) to perform step-by-step reasoning on complex math-related tasks. However, most existing math reasoning datasets may be not able to fully evaluate and analyze the ability of LLMs in manipulating tools and performing reasoning, as they may only require very few invocations of tools or miss annotations for evaluating intermediate reasoning steps. To address the issue, we construct \textbf{CARP}, a new Chinese dataset consisting of 4,886 computation-intensive algebra problems with formulated annotations on intermediate steps. In CARP, we test four LLMs with CoT prompting, and find that they are all prone to make mistakes at the early steps of the solution, leading to wrong answers. Based on this finding, we propose a new approach that can deliberate the reasoning steps with tool interfaces, namely \textbf{DELI}. In DELI, we first initialize a step-by-step solution based on retrieved exemplars, then iterate two deliberation procedures that check and refine the intermediate steps of the generated solution, from the perspectives of tool manipulation and natural language reasoning, until obtaining converged solutions or reaching the maximum turn. Experimental results on CARP and six other datasets show that the proposed DELI mostly outperforms competitive baselines, and can further boost the performance of existing CoT methods. Our data and code are available in \url{https://github.com/RUCAIBox/CARP}.
comment: 17 pages, working in progress
☆ An Information-Theoretic Analysis of Self-supervised Discrete Representations of Speech
Self-supervised representation learning for speech often involves a quantization step that transforms the acoustic input into discrete units. However, it remains unclear how to characterize the relationship between these discrete units and abstract phonetic categories such as phonemes. In this paper, we develop an information-theoretic framework whereby we represent each phonetic category as a distribution over discrete units. We then apply our framework to two different self-supervised models (namely wav2vec 2.0 and XLSR) and use American English speech as a case study. Our study demonstrates that the entropy of phonetic distributions reflects the variability of the underlying speech sounds, with phonetically similar sounds exhibiting similar distributions. While our study confirms the lack of direct, one-to-one correspondence, we find an intriguing, indirect relationship between phonetic categories and discrete units.
comment: Accepted in Interspeech 2023
☆ Commonsense Knowledge Transfer for Pre-trained Language Models ACL 2023
Despite serving as the foundation models for a wide range of NLP benchmarks, pre-trained language models have shown limited capabilities of acquiring implicit commonsense knowledge from self-supervision alone, compared to learning linguistic and factual knowledge that appear more explicitly in the surface patterns in text. In this work, we introduce commonsense knowledge transfer, a framework to transfer the commonsense knowledge stored in a neural commonsense knowledge model to a general-purpose pre-trained language model. It first exploits general texts to form queries for extracting commonsense knowledge from the neural commonsense knowledge model and then refines the language model with two self-supervised objectives: commonsense mask infilling and commonsense relation prediction, which align human language with the underlying commonsense knowledge. Empirical results show that our approach consistently improves the model's performance on downstream tasks that require commonsense reasoning. Moreover, we find that the improvement is more significant in the few-shot setting. This suggests that our approach helps language models better transfer to downstream tasks without extensive supervision by injecting commonsense knowledge into their parameters.
comment: ACL 2023 Findings
☆ Evolution of Efficient Symbolic Communication Codes
The paper explores how the human natural language structure can be seen as a product of evolution of inter-personal communication code, targeting maximisation of such culture-agnostic and cross-lingual metrics such as anti-entropy, compression factor and cross-split F1 score. The exploration is done as part of a larger unsupervised language learning effort, the attempt is made to perform meta-learning in a space of hyper-parameters maximising F1 score based on the "ground truth" language structure, by means of maximising the metrics mentioned above. The paper presents preliminary results of cross-lingual word-level segmentation tokenisation study for Russian, Chinese and English as well as subword segmentation or morphological parsing study for English. It is found that language structure form the word-level segmentation or tokenisation can be found as driven by all of these metrics, anti-entropy being more relevant to English and Russian while compression factor more specific for Chinese. The study for subword segmentation or morphological parsing on English lexicon has revealed straight connection between the compression been found to be associated with compression factor, while, surprising, the same connection with anti-entropy has turned to be the inverse.
comment: 9 pages, 6 figures
☆ Modular Transformers: Compressing Transformers into Modularized Layers for Flexible Efficient Inference ACL 2023
Pre-trained Transformer models like T5 and BART have advanced the state of the art on a wide range of text generation tasks. Compressing these models into smaller ones has become critically important for practical use. Common neural network compression techniques such as knowledge distillation or quantization are limited to static compression where the compression ratio is fixed. In this paper, we introduce Modular Transformers, a modularized encoder-decoder framework for flexible sequence-to-sequence model compression. Modular Transformers train modularized layers that have the same function of two or more consecutive layers in the original model via module replacing and knowledge distillation. After training, the modularized layers can be flexibly assembled into sequence-to-sequence models that meet different performance-efficiency trade-offs. Experimental results show that after a single training phase, by simply varying the assembling strategy, Modular Transformers can achieve flexible compression ratios from 1.1x to 6x with little to moderate relative performance drop.
comment: ACL 2023 Findings
☆ "Are you telling me to put glasses on the dog?'' Content-Grounded Annotation of Instruction Clarification Requests in the CoDraw Dataset
Instruction Clarification Requests are a mechanism to solve communication problems, which is very functional in instruction-following interactions. Recent work has argued that the CoDraw dataset is a valuable source of naturally occurring iCRs. Beyond identifying when iCRs should be made, dialogue models should also be able to generate them with suitable form and content. In this work, we introduce CoDraw-iCR (v2), which extends the existing iCR identifiers fine-grained information grounded in the underlying dialogue game items and possible actions. Our annotation can serve to model and evaluate repair capabilities of dialogue agents.
comment: Work in progress
☆ bgGLUE: A Bulgarian General Language Understanding Evaluation Benchmark ACL 2023
We present bgGLUE (Bulgarian General Language Understanding Evaluation), a benchmark for evaluating language models on Natural Language Understanding (NLU) tasks in Bulgarian. Our benchmark includes NLU tasks targeting a variety of NLP problems (e.g., natural language inference, fact-checking, named entity recognition, sentiment analysis, question answering, etc.) and machine learning tasks (sequence labeling, document-level classification, and regression). We run the first systematic evaluation of pre-trained language models for Bulgarian, comparing and contrasting results across the nine tasks in the benchmark. The evaluation results show strong performance on sequence labeling tasks, but there is a lot of room for improvement for tasks that require more complex reasoning. We make bgGLUE publicly available together with the fine-tuning and the evaluation code, as well as a public leaderboard at https://bgglue.github.io/, and we hope that it will enable further advancements in developing NLU models for Bulgarian.
comment: Accepted to ACL 2023 (Main Conference)
☆ Leverage Points in Modality Shifts: Comparing Language-only and Multimodal Word Representations
Multimodal embeddings aim to enrich the semantic information in neural representations of language compared to text-only models. While different embeddings exhibit different applicability and performance on downstream tasks, little is known about the systematic representation differences attributed to the visual modality. Our paper compares word embeddings from three vision-and-language models (CLIP, OpenCLIP and Multilingual CLIP) and three text-only models, with static (FastText) as well as contextual representations (multilingual BERT; XLM-RoBERTa). This is the first large-scale study of the effect of visual grounding on language representations, including 46 semantic parameters. We identify meaning properties and relations that characterize words whose embeddings are most affected by the inclusion of visual modality in the training data; that is, points where visual grounding turns out most important. We find that the effect of visual modality correlates most with denotational semantic properties related to concreteness, but is also detected for several specific semantic classes, as well as for valence, a sentiment-related connotational property of linguistic expressions.
comment: Accepted for StarSEM 2023
☆ Long Text Generation Challenge
We propose a shared task of human-like long text generation, LTG Challenge, that asks models to output a consistent human-like long text (a Harry Potter generic audience fanfic in English), given a prompt of about 1000 tokens. We suggest a novel statistical metric of the text structuredness, GloVe Autocorrelations Power/ Exponential Law Mean Absolute Percentage Error Ratio (GAPELMAPER) and a human evaluation protocol. We hope that LTG can open new avenues for researchers to investigate sampling approaches, prompting strategies, autoregressive and non-autoregressive text generation architectures and break the barrier to generate consistent long (40K+ token) texts.
comment: Submitted to INLG 2023
☆ Arbitrary Few Parameters are Good Enough for Adapting Large-scale Pre-trained Language Models
Parameter-efficient tuning (PET) methods can effectively drive extremely large pre-trained language models (PLMs) by only training minimal parameters. Different PET methods utilize different manually designed modules. In a small PLM, there are usually noticeable performance differences among PET methods. Nevertheless, when a PLM's scale grows up to tens of billions of parameters, all PET methods achieve almost the same performance and even perform on par with the full-parameter fine-tuning method. Hence, we hypothesize that model scaling can mitigate the design differences (the module structures and the number of trainable parameters) among PET methods. To study this hypothesis, we introduce a more flexible PET method - arbitrary PET (APET) method - to be compatible with arbitrary module structures and any number of trainable parameters. Then, we experiment on $11$ NLP tasks of $5$ types and $2$ representative PLMs. From our investigations, we find that the model scaling (1) mitigates the effects of the arbitrary module structure on the performance of tuning methods, and (2) enables the tuning methods to optimize fewer parameters to achieve the full-parameter fine-tuning performance. Intriguingly, we also observe that all tuning methods require almost the same number of trainable parameters to drive PLMs. We discuss this phenomenon and the above two findings collectively from optimization perspectives to fathom the mechanisms behind them. These conclusions not only demonstrate the positive impact of model scaling on tuning methods but disclose its mechanisms, which help us design more effective and efficient tuning methods on larger-scale PLMs.
☆ SpellMapper: A non-autoregressive neural spellchecker for ASR customization with candidate retrieval based on n-gram mappings INTERSPEECH 2023
Contextual spelling correction models are an alternative to shallow fusion to improve automatic speech recognition (ASR) quality given user vocabulary. To deal with large user vocabularies, most of these models include candidate retrieval mechanisms, usually based on minimum edit distance between fragments of ASR hypothesis and user phrases. However, the edit-distance approach is slow, non-trainable, and may have low recall as it relies only on common letters. We propose: 1) a novel algorithm for candidate retrieval, based on misspelled n-gram mappings, which gives up to 90% recall with just the top 10 candidates on Spoken Wikipedia; 2) a non-autoregressive neural model based on BERT architecture, where the initial transcript and ten candidates are combined into one input. The experiments on Spoken Wikipedia show 21.4% word error rate improvement compared to a baseline ASR system.
comment: Accepted by INTERSPEECH 2023
☆ Finding the SWEET Spot: Analysis and Improvement of Adaptive Inference in Low Resource Settings ACL 2023
Adaptive inference is a simple method for reducing inference costs. The method works by maintaining multiple classifiers of different capacities, and allocating resources to each test instance according to its difficulty. In this work, we compare the two main approaches for adaptive inference, Early-Exit and Multi-Model, when training data is limited. First, we observe that for models with the same architecture and size, individual Multi-Model classifiers outperform their Early-Exit counterparts by an average of 2.3%. We show that this gap is caused by Early-Exit classifiers sharing model parameters during training, resulting in conflicting gradient updates of model weights. We find that despite this gap, Early-Exit still provides a better speed-accuracy trade-off due to the overhead of the Multi-Model approach. To address these issues, we propose SWEET (Separating Weights in Early Exit Transformers), an Early-Exit fine-tuning method that assigns each classifier its own set of unique model weights, not updated by other classifiers. We compare SWEET's speed-accuracy curve to standard Early-Exit and Multi-Model baselines and find that it outperforms both methods at fast speeds while maintaining comparable scores to Early-Exit at slow speeds. Moreover, SWEET individual classifiers outperform Early-Exit ones by 1.1% on average. SWEET enjoys the benefits of both methods, paving the way for further reduction of inference costs in NLP.
comment: Proceedings of ACL 2023
☆ Does Character-level Information Always Improve DRS-based Semantic Parsing? ACL2023
Even in the era of massive language models, it has been suggested that character-level representations improve the performance of neural models. The state-of-the-art neural semantic parser for Discourse Representation Structures uses character-level representations, improving performance in the four languages (i.e., English, German, Dutch, and Italian) in the Parallel Meaning Bank dataset. However, how and why character-level information improves the parser's performance remains unclear. This study provides an in-depth analysis of performance changes by order of character sequences. In the experiments, we compare F1-scores by shuffling the order and randomizing character sequences after testing the performance of character-level information. Our results indicate that incorporating character-level information does not improve the performance in English and German. In addition, we find that the parser is not sensitive to correct character order in Dutch. Nevertheless, performance improvements are observed when using character-level information.
comment: 10 pages. To appear in the 12th Joint Conference on Lexical and Computational Semantics (*SEM 2023) with ACL2023
☆ A Mathematical Abstraction for Balancing the Trade-off Between Creativity and Reality in Large Language Models
Large Language Models have become popular for their remarkable capabilities in human-oriented tasks and traditional natural language processing tasks. Its efficient functioning is attributed to the attention mechanism in the Transformer architecture, enabling it to concentrate on particular aspects of the input. LLMs are increasingly being used in domains such as generating prose, poetry or art, which require the model to be creative (e.g. Adobe firefly). LLMs possess advanced language generation abilities that enable them to generate distinctive and captivating content. This utilization of LLMs in generating narratives shows their flexibility and potential for use in domains that extend beyond conventional natural language processing duties. In different contexts, we may expect the LLM to generate factually correct answers, that match reality; e.g., question-answering systems or online assistants. In such situations, being correct is critical to LLMs being trusted in practice. The Bing Chatbot provides its users with the flexibility to select one of the three output modes: creative, balanced, and precise. Each mode emphasizes creativity and factual accuracy differently. In this work, we provide a mathematical abstraction to describe creativity and reality based on certain losses. A model trained on these losses balances the trade-off between the creativity and reality of the model.
☆ Exposing Bias in Online Communities through Large-Scale Language Models
Progress in natural language generation research has been shaped by the ever-growing size of language models. While large language models pre-trained on web data can generate human-sounding text, they also reproduce social biases and contribute to the propagation of harmful stereotypes. This work utilises the flaw of bias in language models to explore the biases of six different online communities. In order to get an insight into the communities' viewpoints, we fine-tune GPT-Neo 1.3B with six social media datasets. The bias of the resulting models is evaluated by prompting the models with different demographics and comparing the sentiment and toxicity values of these generations. Together, these methods reveal that bias differs in type and intensity for the various models. This work not only affirms how easily bias is absorbed from training data but also presents a scalable method to identify and compare the bias of different datasets or communities. Additionally, the examples generated for this work demonstrate the limitations of using automated sentiment and toxicity classifiers in bias research.
☆ Exploring and Verbalizing Academic Ideas by Concept Co-occurrence ACL 2023
Researchers usually come up with new ideas only after thoroughly comprehending vast quantities of literature. The difficulty of this procedure is exacerbated by the fact that the number of academic publications is growing exponentially. In this study, we devise a framework based on concept co-occurrence for academic idea inspiration, which has been integrated into a research assistant system. From our perspective, the fusion of two concepts that co-occur in an academic paper can be regarded as an important way of the emergence of a new idea. We construct evolving concept graphs according to the co-occurrence relationship of concepts from 20 disciplines or topics. Then we design a temporal link prediction method based on masked language model to explore potential connections between different concepts. To verbalize the newly discovered connections, we also utilize the pretrained language model to generate a description of an idea based on a new data structure called co-occurrence citation quintuple. We evaluate our proposed system using both automatic metrics and human assessment. The results demonstrate that our system has broad prospects and can assist researchers in expediting the process of discovering new ideas.
comment: Accepted by ACL 2023
☆ End-to-End Joint Target and Non-Target Speakers ASR
This paper proposes a novel automatic speech recognition (ASR) system that can transcribe individual speaker's speech while identifying whether they are target or non-target speakers from multi-talker overlapped speech. Target-speaker ASR systems are a promising way to only transcribe a target speaker's speech by enrolling the target speaker's information. However, in conversational ASR applications, transcribing both the target speaker's speech and non-target speakers' ones is often required to understand interactive information. To naturally consider both target and non-target speakers in a single ASR model, our idea is to extend autoregressive modeling-based multi-talker ASR systems to utilize the enrollment speech of the target speaker. Our proposed ASR is performed by recursively generating both textual tokens and tokens that represent target or non-target speakers. Our experiments demonstrate the effectiveness of our proposed method.
comment: Accepted at Interspeech 2023
☆ OWQ: Lessons learned from activation outliers for weight quantization in large language models
Large language models (LLMs) with hundreds of billions of parameters show impressive results across various language tasks using simple prompt tuning and few-shot examples, without the need for task-specific fine-tuning. However, their enormous size requires multiple server-grade GPUs even for inference, creating a significant cost barrier. To address this limitation, we introduce a novel post-training quantization method for weights with minimal quality degradation. While activation outliers are known to be problematic in activation quantization, our theoretical analysis suggests that we can identify factors contributing to weight quantization errors by considering activation outliers. We propose an innovative PTQ scheme called outlier-aware weight quantization (OWQ), which identifies vulnerable weights and allocates high-precision to them. Our extensive experiments demonstrate that the 3.01-bit models produced by OWQ exhibit comparable quality to the 4-bit models generated by OPTQ.
☆ Probing Physical Reasoning with Counter-Commonsense Context ACL 2023
In this study, we create a CConS (Counter-commonsense Contextual Size comparison) dataset to investigate how physical commonsense affects the contextualized size comparison task; the proposed dataset consists of both contexts that fit physical commonsense and those that do not. This dataset tests the ability of language models to predict the size relationship between objects under various contexts generated from our curated noun list and templates. We measure the ability of several masked language models and generative models. The results show that while large language models can use prepositions such as ``in'' and ``into'' in the provided context to infer size relationships, they fail to use verbs and thus make incorrect judgments led by their prior physical commonsense.
comment: Accepted to ACL 2023(Short Paper)
☆ A Technical Report for Polyglot-Ko: Open-Source Large-Scale Korean Language Models
Polyglot is a pioneering project aimed at enhancing the non-English language performance of multilingual language models. Despite the availability of various multilingual models such as mBERT (Devlin et al., 2019), XGLM (Lin et al., 2022), and BLOOM (Scao et al., 2022), researchers and developers often resort to building monolingual models in their respective languages due to the dissatisfaction with the current multilingual models non-English language capabilities. Addressing this gap, we seek to develop advanced multilingual language models that offer improved performance in non-English languages. In this paper, we introduce the Polyglot Korean models, which represent a specific focus rather than being multilingual in nature. In collaboration with TUNiB, our team collected 1.2TB of Korean data meticulously curated for our research journey. We made a deliberate decision to prioritize the development of Korean models before venturing into multilingual models. This choice was motivated by multiple factors: firstly, the Korean models facilitated performance comparisons with existing multilingual models; and finally, they catered to the specific needs of Korean companies and researchers. This paper presents our work in developing the Polyglot Korean models, which propose some steps towards addressing the non-English language performance gap in multilingual language models.
☆ Large Language Model Augmented Narrative Driven Recommendations
Narrative-driven recommendation (NDR) presents an information access problem where users solicit recommendations with verbose descriptions of their preferences and context, for example, travelers soliciting recommendations for points of interest while describing their likes/dislikes and travel circumstances. These requests are increasingly important with the rise of natural language-based conversational interfaces for search and recommendation systems. However, NDR lacks abundant training data for models, and current platforms commonly do not support these requests. Fortunately, classical user-item interaction datasets contain rich textual data, e.g., reviews, which often describe user preferences and context - this may be used to bootstrap training for NDR models. In this work, we explore using large language models (LLMs) for data augmentation to train NDR models. We use LLMs for authoring synthetic narrative queries from user-item interactions with few-shot prompting and train retrieval models for NDR on synthetic queries and user-item interaction data. Our experiments demonstrate that this is an effective strategy for training small-parameter retrieval models that outperform other retrieval and LLM baselines for narrative-driven recommendation.
comment: Pre-print
☆ Sen2Pro: A Probabilistic Perspective to Sentence Embedding from Pre-trained Language Model ACL2023
Sentence embedding is one of the most fundamental tasks in Natural Language Processing and plays an important role in various tasks. The recent breakthrough in sentence embedding is achieved by pre-trained language models (PLMs). Despite its success, an embedded vector (Sen2Vec) representing a point estimate does not naturally express uncertainty in a taskagnostic way. This paper thereby proposes an efficient framework on probabilistic sentence embedding (Sen2Pro) from PLMs, and it represents a sentence as a probability density distribution in an embedding space to reflect both model uncertainty and data uncertainty (i.e., many-to-one nature) in the sentence representation. The proposed framework performs in a plug-and-play way without retraining PLMs anymore, and it is easy to implement and generally applied on top of any PLM. The superiority of Sen2Pro over Sen2Vec has been theoretically verified and practically illustrated on different NLP tasks.
comment: Accepted to ACL2023 workshop Rep4NLP
☆ Extract and Attend: Improving Entity Translation in Neural Machine Translation
While Neural Machine Translation(NMT) has achieved great progress in recent years, it still suffers from inaccurate translation of entities (e.g., person/organization name, location), due to the lack of entity training instances. When we humans encounter an unknown entity during translation, we usually first look up in a dictionary and then organize the entity translation together with the translations of other parts to form a smooth target sentence. Inspired by this translation process, we propose an Extract-and-Attend approach to enhance entity translation in NMT, where the translation candidates of source entities are first extracted from a dictionary and then attended to by the NMT model to generate the target sentence. Specifically, the translation candidates are extracted by first detecting the entities in a source sentence and then translating the entities through looking up in a dictionary. Then, the extracted candidates are added as a prefix of the decoder input to be attended to by the decoder when generating the target sentence through self-attention. Experiments conducted on En-Zh and En-Ru demonstrate that the proposed method is effective on improving both the translation accuracy of entities and the overall translation quality, with up to 35% reduction on entity error rate and 0.85 gain on BLEU and 13.8 gain on COMET.
☆ Fine-Tuning Language Models with Advantage-Induced Policy Alignment
Reinforcement learning from human feedback (RLHF) has emerged as a reliable approach to aligning large language models (LLMs) to human preferences. Among the plethora of RLHF techniques, proximal policy optimization (PPO) is of the most widely used methods. Despite its popularity, however, PPO may suffer from mode collapse, instability, and poor sample efficiency. We show that these issues can be alleviated by a novel algorithm that we refer to as Advantage-Induced Policy Alignment (APA), which leverages a squared error loss function based on the estimated advantages. We demonstrate empirically that APA consistently outperforms PPO in language tasks by a large margin, when a separate reward model is employed as the evaluator. In addition, compared with PPO, APA offers a more stable form of control over the deviation from the model's initial policy, ensuring that the model improves its performance without collapsing to deterministic output. In addition to empirical results, we also provide a theoretical justification supporting the design of our loss function.
♻ ☆ Paraphrastic Representations at Scale EMNLP 2022
We present a system that allows users to train their own state-of-the-art paraphrastic sentence representations in a variety of languages. We also release trained models for English, Arabic, German, French, Spanish, Russian, Turkish, and Chinese. We train these models on large amounts of data, achieving significantly improved performance from the original papers proposing the methods on a suite of monolingual semantic similarity, cross-lingual semantic similarity, and bitext mining tasks. Moreover, the resulting models surpass all prior work on unsupervised semantic textual similarity, significantly outperforming even BERT-based models like Sentence-BERT (Reimers and Gurevych, 2019). Additionally, our models are orders of magnitude faster than prior work and can be used on CPU with little difference in inference speed (even improved speed over GPU when using more CPU cores), making these models an attractive choice for users without access to GPUs or for use on embedded devices. Finally, we add significantly increased functionality to the code bases for training paraphrastic sentence models, easing their use for both inference and for training them for any desired language with parallel data. We also include code to automatically download and preprocess training data.
comment: Published as a demo paper at EMNLP 2022
♻ ☆ Beyond Contrastive Learning: A Variational Generative Model for Multilingual Retrieval ACL 2023
Contrastive learning has been successfully used for retrieval of semantically aligned sentences, but it often requires large batch sizes or careful engineering to work well. In this paper, we instead propose a generative model for learning multilingual text embeddings which can be used to retrieve or score sentence pairs. Our model operates on parallel data in $N$ languages and, through an approximation we introduce, efficiently encourages source separation in this multilingual setting, separating semantic information that is shared between translations from stylistic or language-specific variation. We show careful large-scale comparisons between contrastive and generation-based approaches for learning multilingual text embeddings, a comparison that has not been done to the best of our knowledge despite the popularity of these approaches. We evaluate this method on a suite of tasks including semantic similarity, bitext mining, and cross-lingual question retrieval -- the last of which we introduce in this paper. Overall, our Variational Multilingual Source-Separation Transformer (VMSST) model outperforms both a strong contrastive and generative baseline on these tasks.
comment: Published as a long paper at ACL 2023
♻ ☆ Evaluating natural language processing models with generalization metrics that do not need access to any training or testing data
Selecting suitable architecture parameters and training hyperparameters is essential for enhancing machine learning (ML) model performance. Several recent empirical studies conduct large-scale correlational analysis on neural networks (NNs) to search for effective \emph{generalization metrics} that can guide this type of model selection. Effective metrics are typically expected to correlate strongly with test performance. In this paper, we expand on prior analyses by examining generalization-metric-based model selection with the following objectives: (i) focusing on natural language processing (NLP) tasks, as prior work primarily concentrates on computer vision (CV) tasks; (ii) considering metrics that directly predict \emph{test error} instead of the \emph{generalization gap}; (iii) exploring metrics that do not need access to data to compute. From these objectives, we are able to provide the first model selection results on large pretrained Transformers from Huggingface using generalization metrics. Our analyses consider (I) hundreds of Transformers trained in different settings, in which we systematically vary the amount of data, the model size and the optimization hyperparameters, (II) a total of 51 pretrained Transformers from eight families of Huggingface NLP models, including GPT2, BERT, etc., and (III) a total of 28 existing and novel generalization metrics. Despite their niche status, we find that metrics derived from the heavy-tail (HT) perspective are particularly useful in NLP tasks, exhibiting stronger correlations than other, more popular metrics. To further examine these metrics, we extend prior formulations relying on power law (PL) spectral distributions to exponential (EXP) and exponentially-truncated power law (E-TPL) families.
♻ ☆ Heterformer: Transformer-based Deep Node Representation Learning on Heterogeneous Text-Rich Networks KDD 2023
Representation learning on networks aims to derive a meaningful vector representation for each node, thereby facilitating downstream tasks such as link prediction, node classification, and node clustering. In heterogeneous text-rich networks, this task is more challenging due to (1) presence or absence of text: Some nodes are associated with rich textual information, while others are not; (2) diversity of types: Nodes and edges of multiple types form a heterogeneous network structure. As pretrained language models (PLMs) have demonstrated their effectiveness in obtaining widely generalizable text representations, a substantial amount of effort has been made to incorporate PLMs into representation learning on text-rich networks. However, few of them can jointly consider heterogeneous structure (network) information as well as rich textual semantic information of each node effectively. In this paper, we propose Heterformer, a Heterogeneous Network-Empowered Transformer that performs contextualized text encoding and heterogeneous structure encoding in a unified model. Specifically, we inject heterogeneous structure information into each Transformer layer when encoding node texts. Meanwhile, Heterformer is capable of characterizing node/edge type heterogeneity and encoding nodes with or without texts. We conduct comprehensive experiments on three tasks (i.e., link prediction, node classification, and node clustering) on three large-scale datasets from different domains, where Heterformer outperforms competitive baselines significantly and consistently.
comment: KDD 2023. (Code: https://github.com/PeterGriffinJin/Heterformer)
♻ ☆ On Second Thought, Let's Not Think Step by Step! Bias and Toxicity in Zero-Shot Reasoning ACL 2023
Generating a Chain of Thought (CoT) has been shown to consistently improve large language model (LLM) performance on a wide range of NLP tasks. However, prior work has mainly focused on logical reasoning tasks (e.g. arithmetic, commonsense QA); it remains unclear whether improvements hold for more diverse types of reasoning, especially in socially situated contexts. Concretely, we perform a controlled evaluation of zero-shot CoT across two socially sensitive domains: harmful questions and stereotype benchmarks. We find that zero-shot CoT reasoning in sensitive domains significantly increases a model's likelihood to produce harmful or undesirable output, with trends holding across different prompt formats and model variants. Furthermore, we show that harmful CoTs increase with model size, but decrease with improved instruction following. Our work suggests that zero-shot CoT should be used with caution on socially important tasks, especially when marginalized groups or sensitive topics are involved.
comment: ACL 2023 Main Conference
♻ ☆ Learning to Simulate Natural Language Feedback for Interactive Semantic Parsing ACL 2023
Interactive semantic parsing based on natural language (NL) feedback, where users provide feedback to correct the parser mistakes, has emerged as a more practical scenario than the traditional one-shot semantic parsing. However, prior work has heavily relied on human-annotated feedback data to train the interactive semantic parser, which is prohibitively expensive and not scalable. In this work, we propose a new task of simulating NL feedback for interactive semantic parsing. We accompany the task with a novel feedback evaluator. The evaluator is specifically designed to assess the quality of the simulated feedback, based on which we decide the best feedback simulator from our proposed variants. On a text-to-SQL dataset, we show that our feedback simulator can generate high-quality NL feedback to boost the error correction ability of a specific parser. In low-data settings, our feedback simulator can help achieve comparable error correction performance as trained using the costly, full set of human annotations.
comment: Accepted to ACL 2023. 18 pages, 6 figures
♻ ☆ Investigating Massive Multilingual Pre-Trained Machine Translation Models for Clinical Domain via Transfer Learning ACL-2023
Massively multilingual pre-trained language models (MMPLMs) are developed in recent years demonstrating superpowers and the pre-knowledge they acquire for downstream tasks. This work investigates whether MMPLMs can be applied to clinical domain machine translation (MT) towards entirely unseen languages via transfer learning. We carry out an experimental investigation using Meta-AI's MMPLMs ``wmt21-dense-24-wide-en-X and X-en (WMT21fb)'' which were pre-trained on 7 language pairs and 14 translation directions including English to Czech, German, Hausa, Icelandic, Japanese, Russian, and Chinese, and the opposite direction. We fine-tune these MMPLMs towards English-\textit{Spanish} language pair which \textit{did not exist at all} in their original pre-trained corpora both implicitly and explicitly. We prepare carefully aligned \textit{clinical} domain data for this fine-tuning, which is different from their original mixed domain knowledge. Our experimental result shows that the fine-tuning is very successful using just 250k well-aligned in-domain EN-ES segments for three sub-task translation testings: clinical cases, clinical terms, and ontology concepts. It achieves very close evaluation scores to another MMPLM NLLB from Meta-AI, which included Spanish as a high-resource setting in the pre-training. To the best of our knowledge, this is the first work on using MMPLMs towards \textit{clinical domain transfer-learning NMT} successfully for totally unseen languages during pre-training.
comment: Accepted to ClinicalNLP-2023 WS@ACL-2023
♻ ☆ Are We Really Making Much Progress in Text Classification? A Comparative Review ACL
This study reviews and compares methods for single-label and multi-label text classification, categorized into bag-of-words, sequence-based, graph-based, and hierarchical methods. The comparison aggregates results from the literature over five single-label and seven multi-label datasets and complements them with new experiments. The findings reveal that all recently proposed graph-based and hierarchy-based methods fail to outperform pre-trained language models and sometimes perform worse than standard machine learning methods like a multilayer perceptron on a bag-of-words. To assess the true scientific progress in text classification, future work should thoroughly test against strong bag-of-words baselines and state-of-the-art pre-trained language models.
comment: Update: revised text and new included methods. This work is an extension of "Bag-of-Words vs. Graph vs. Sequence in Text Classification: Questioning the Necessity of Text-Graphs and the Surprising Strength of a Wide MLP. ACL (1) 2022: 4038-4051", URL: https://aclanthology.org/2022.acl-long.279/
♻ ☆ Improving Self-training for Cross-lingual Named Entity Recognition with Contrastive and Prototype Learning ACL2023
In cross-lingual named entity recognition (NER), self-training is commonly used to bridge the linguistic gap by training on pseudo-labeled target-language data. However, due to sub-optimal performance on target languages, the pseudo labels are often noisy and limit the overall performance. In this work, we aim to improve self-training for cross-lingual NER by combining representation learning and pseudo label refinement in one coherent framework. Our proposed method, namely ContProto mainly comprises two components: (1) contrastive self-training and (2) prototype-based pseudo-labeling. Our contrastive self-training facilitates span classification by separating clusters of different classes, and enhances cross-lingual transferability by producing closely-aligned representations between the source and target language. Meanwhile, prototype-based pseudo-labeling effectively improves the accuracy of pseudo labels during training. We evaluate ContProto on multiple transfer pairs, and experimental results show our method brings in substantial improvements over current state-of-the-art methods.
comment: Accepted by ACL2023
♻ ☆ Modular and On-demand Bias Mitigation with Attribute-Removal Subnetworks ACL 2023
Societal biases are reflected in large pre-trained language models and their fine-tuned versions on downstream tasks. Common in-processing bias mitigation approaches, such as adversarial training and mutual information removal, introduce additional optimization criteria, and update the model to reach a new debiased state. However, in practice, end-users and practitioners might prefer to switch back to the original model, or apply debiasing only on a specific subset of protected attributes. To enable this, we propose a novel modular bias mitigation approach, consisting of stand-alone highly sparse debiasing subnetworks, where each debiasing module can be integrated into the core model on-demand at inference time. Our approach draws from the concept of \emph{diff} pruning, and proposes a novel training regime adaptable to various representation disentanglement optimizations. We conduct experiments on three classification tasks with gender, race, and age as protected attributes. The results show that our modular approach, while maintaining task performance, improves (or at least remains on-par with) the effectiveness of bias mitigation in comparison with baseline finetuning. Particularly on a two-attribute dataset, our approach with separately learned debiasing subnetworks shows effective utilization of either or both the subnetworks for selective bias mitigation.
comment: Accepted in Findings of ACL 2023
♻ ☆ Learning to Imagine: Visually-Augmented Natural Language Generation ACL 2023
People often imagine relevant scenes to aid in the writing process. In this work, we aim to utilize visual information for composition in the same manner as humans. We propose a method, LIVE, that makes pre-trained language models (PLMs) Learn to Imagine for Visuallyaugmented natural language gEneration. First, we imagine the scene based on the text: we use a diffusion model to synthesize high-quality images conditioned on the input texts. Second, we use CLIP to determine whether the text can evoke the imagination in a posterior way. Finally, our imagination is dynamic, and we conduct synthesis for each sentence rather than generate only one image for an entire paragraph. Technically, we propose a novel plug-and-play fusion layer to obtain visually-augmented representations for each text. Our vision-text fusion layer is compatible with Transformerbased architecture. We have conducted extensive experiments on four generation tasks using BART and T5, and the automatic results and human evaluation demonstrate the effectiveness of our proposed method. We will release the code, model, and data at the link: https://github.com/RUCAIBox/LIVE.
comment: Accepted by ACL 2023
♻ ☆ Constructing Code-mixed Universal Dependency Forest for Unbiased Cross-lingual Relation Extraction
Latest efforts on cross-lingual relation extraction (XRE) aggressively leverage the language-consistent structural features from the universal dependency (UD) resource, while they may largely suffer from biased transfer (e.g., either target-biased or source-biased) due to the inevitable linguistic disparity between languages. In this work, we investigate an unbiased UD-based XRE transfer by constructing a type of code-mixed UD forest. We first translate the sentence of the source language to the parallel target-side language, for both of which we parse the UD tree respectively. Then, we merge the source-/target-side UD structures as a unified code-mixed UD forest. With such forest features, the gaps of UD-based XRE between the training and predicting phases can be effectively closed. We conduct experiments on the ACE XRE benchmark datasets, where the results demonstrate that the proposed code-mixed UD forests help unbiased UD-based XRE transfer, with which we achieve significant XRE performance gains.
♻ ☆ UnifieR: A Unified Retriever for Large-Scale Retrieval KDD
Large-scale retrieval is to recall relevant documents from a huge collection given a query. It relies on representation learning to embed documents and queries into a common semantic encoding space. According to the encoding space, recent retrieval methods based on pre-trained language models (PLM) can be coarsely categorized into either dense-vector or lexicon-based paradigms. These two paradigms unveil the PLMs' representation capability in different granularities, i.e., global sequence-level compression and local word-level contexts, respectively. Inspired by their complementary global-local contextualization and distinct representing views, we propose a new learning framework, UnifieR which unifies dense-vector and lexicon-based retrieval in one model with a dual-representing capability. Experiments on passage retrieval benchmarks verify its effectiveness in both paradigms. A uni-retrieval scheme is further presented with even better retrieval quality. We lastly evaluate the model on BEIR benchmark to verify its transferability.
comment: To appear at KDD ADS 2023
♻ ☆ Target-Side Augmentation for Document-Level Machine Translation ACL2023
Document-level machine translation faces the challenge of data sparsity due to its long input length and a small amount of training data, increasing the risk of learning spurious patterns. To address this challenge, we propose a target-side augmentation method, introducing a data augmentation (DA) model to generate many potential translations for each source document. Learning on these wider range translations, an MT model can learn a smoothed distribution, thereby reducing the risk of data sparsity. We demonstrate that the DA model, which estimates the posterior distribution, largely improves the MT performance, outperforming the previous best system by 2.30 s-BLEU on News and achieving new state-of-the-art on News and Europarl benchmarks. Our code is available at https://github.com/baoguangsheng/target-side-augmentation.
comment: Accepted by ACL2023 main conference
♻ ☆ On the Copying Problem of Unsupervised NMT: A Training Schedule with a Language Discriminator Loss
Although unsupervised neural machine translation (UNMT) has achieved success in many language pairs, the copying problem, i.e., directly copying some parts of the input sentence as the translation, is common among distant language pairs, especially when low-resource languages are involved. We find this issue is closely related to an unexpected copying behavior during online back-translation (BT). In this work, we propose a simple but effective training schedule that incorporates a language discriminator loss. The loss imposes constraints on the intermediate translation so that the translation is in the desired language. By conducting extensive experiments on different language pairs, including similar and distant, high and low-resource languages, we find that our method alleviates the copying problem, thus improving the translation performance on low-resource languages.
comment: IWSLT 2023
♻ ☆ Technical Report on Token Position Bias in Transformers
Language Models (LMs) have shown state-of-the-art performance in Natural Language Processing (NLP) tasks. Downstream tasks such as Named Entity Recognition (NER) or Part-of-Speech (POS) tagging are known to suffer from data imbalance issues, specifically in terms of the ratio of positive to negative examples, and class imbalance. In this paper, we investigate an additional specific issue for language models, namely the position bias of positive examples in token classification tasks. Therefore, we conduct an in-depth evaluation of the impact of position bias on the performance of LMs when fine-tuned on Token Classification benchmarks. Our study includes CoNLL03 and OntoNote5.0 for NER, English Tree Bank UD_en and TweeBank for POS tagging. We propose an evaluation approach to investigate position bias in Transformer models. We show that encoders like BERT, ERNIE, ELECTRA, and decoders such as GPT2 and BLOOM can suffer from this bias with an average drop of 3\% and 9\% in their performance. To mitigate this effect, we propose two methods: Random Position Shifting and Context Perturbation, that we apply on batches during the training process. The results show an improvement of $\approx$ 2\% in the performance of the model on CoNLL03, UD_en, and TweeBank.
comment: Updated title of the preprint
♻ ☆ Scan and Snap: Understanding Training Dynamics and Token Composition in 1-layer Transformer
Transformer architecture has shown impressive performance in multiple research domains and has become the backbone of many neural network models. However, there is limited understanding on how it works. In particular, with a simple predictive loss, how the representation emerges from the gradient \emph{training dynamics} remains a mystery. In this paper, for 1-layer transformer with one self-attention layer plus one decoder layer, we analyze its SGD training dynamics for the task of next token prediction in a mathematically rigorous manner. We open the black box of the dynamic process of how the self-attention layer combines input tokens, and reveal the nature of underlying inductive bias. More specifically, with the assumption (a) no positional encoding, (b) long input sequence, and (c) the decoder layer learns faster than the self-attention layer, we prove that self-attention acts as a \emph{discriminative scanning algorithm}: starting from uniform attention, it gradually attends more to distinct key tokens for a specific next token to be predicted, and pays less attention to common key tokens that occur across different next tokens. Among distinct tokens, it progressively drops attention weights, following the order of low to high co-occurrence between the key and the query token in the training set. Interestingly, this procedure does not lead to winner-takes-all, but decelerates due to a \emph{phase transition} that is controllable by the learning rates of the two layers, leaving (almost) fixed token combination. We verify this \textbf{\emph{scan and snap}} dynamics on synthetic and real-world data (WikiText).
comment: Fix minor issues in the proofs and figures. Update figures to reflect the main conclusions more accurately
♻ ☆ SI-LSTM: Speaker Hybrid Long-short Term Memory and Cross Modal Attention for Emotion Recognition in Conversation
Emotion Recognition in Conversation~(ERC) across modalities is of vital importance for a variety of applications, including intelligent healthcare, artificial intelligence for conversation, and opinion mining over chat history. The crux of ERC is to model both cross-modality and cross-time interactions throughout the conversation. Previous methods have made progress in learning the time series information of conversation while lacking the ability to trace down the different emotional states of each speaker in a conversation. In this paper, we propose a recurrent structure called Speaker Information Enhanced Long-Short Term Memory (SI-LSTM) for the ERC task, where the emotional states of the distinct speaker can be tracked in a sequential way to enhance the learning of the emotion in conversation. Further, to improve the learning of multimodal features in ERC, we utilize a cross-modal attention component to fuse the features between different modalities and model the interaction of the important information from different modalities. Experimental results on two benchmark datasets demonstrate the superiority of the proposed SI-LSTM against the state-of-the-art baseline methods in the ERC task on multimodal data.
comment: Modification needed
♻ ☆ A First Look at LLM-Powered Generative News Recommendation
Personalized news recommendation systems have become essential tools for users to navigate the vast amount of online news content, yet existing news recommenders face significant challenges such as the cold-start problem, user profile modeling, and news content understanding. Previous works have typically followed an inflexible routine to address a particular challenge through model design, but are limited in their ability to understand news content and capture user interests. In this paper, we introduce GENRE, an LLM-powered generative news recommendation framework, which leverages pretrained semantic knowledge from large language models to enrich news data. Our aim is to provide a flexible and unified solution for news recommendation by moving from model design to prompt design. We showcase the use of GENRE for personalized news generation, user profiling, and news summarization. Extensive experiments with various popular recommendation models demonstrate the effectiveness of GENRE. We will publish our code and data for other researchers to reproduce our work.
comment: 11 pages
Computer Vision and Pattern Recognition 42
☆ Deep learning powered real-time identification of insects using citizen science data
Insect-pests significantly impact global agricultural productivity and quality. Effective management involves identifying the full insect community, including beneficial insects and harmful pests, to develop and implement integrated pest management strategies. Automated identification of insects under real-world conditions presents several challenges, including differentiating similar-looking species, intra-species dissimilarity and inter-species similarity, several life cycle stages, camouflage, diverse imaging conditions, and variability in insect orientation. A deep-learning model, InsectNet, is proposed to address these challenges. InsectNet is endowed with five key features: (a) utilization of a large dataset of insect images collected through citizen science; (b) label-free self-supervised learning for large models; (c) improving prediction accuracy for species with a small sample size; (d) enhancing model trustworthiness; and (e) democratizing access through streamlined MLOps. This approach allows accurate identification (>96% accuracy) of over 2500 insect species, including pollinator (e.g., butterflies, bees), parasitoid (e.g., some wasps and flies), predator species (e.g., lady beetles, mantises, dragonflies) and harmful pest species (e.g., armyworms, cutworms, grasshoppers, stink bugs). InsectNet can identify invasive species, provide fine-grained insect species identification, and work effectively in challenging backgrounds. It also can abstain from making predictions when uncertain, facilitating seamless human intervention and making it a practical and trustworthy tool. InsectNet can guide citizen science data collection, especially for invasive species where early detection is crucial. Similar approaches may transform other agricultural challenges like disease detection and underscore the importance of data collection, particularly through citizen science efforts..
☆ Systematic Visual Reasoning through Object-Centric Relational Abstraction
Human visual reasoning is characterized by an ability to identify abstract patterns from only a small number of examples, and to systematically generalize those patterns to novel inputs. This capacity depends in large part on our ability to represent complex visual inputs in terms of both objects and relations. Recent work in computer vision has introduced models with the capacity to extract object-centric representations, leading to the ability to process multi-object visual inputs, but falling short of the systematic generalization displayed by human reasoning. Other recent models have employed inductive biases for relational abstraction to achieve systematic generalization of learned abstract rules, but have generally assumed the presence of object-focused inputs. Here, we combine these two approaches, introducing Object-Centric Relational Abstraction (OCRA), a model that extracts explicit representations of both objects and abstract relations, and achieves strong systematic generalization in tasks involving complex visual displays.
☆ Discussion Paper: The Threat of Real Time Deepfakes
Generative deep learning models are able to create realistic audio and video. This technology has been used to impersonate the faces and voices of individuals. These ``deepfakes'' are being used to spread misinformation, enable scams, perform fraud, and blackmail the innocent. The technology continues to advance and today attackers have the ability to generate deepfakes in real-time. This new capability poses a significant threat to society as attackers begin to exploit the technology in advances social engineering attacks. In this paper, we discuss the implications of this emerging threat, identify the challenges with preventing these attacks and suggest a better direction for researching stronger defences.
☆ Multi-Predict: Few Shot Predictors For Efficient Neural Architecture Search
Many hardware-aware neural architecture search (NAS) methods have been developed to optimize the topology of neural networks (NN) with the joint objectives of higher accuracy and lower latency. Recently, both accuracy and latency predictors have been used in NAS with great success, achieving high sample efficiency and accurate modeling of hardware (HW) device latency respectively. However, a new accuracy predictor needs to be trained for every new NAS search space or NN task, and a new latency predictor needs to be additionally trained for every new HW device. In this paper, we explore methods to enable multi-task, multi-search-space, and multi-HW adaptation of accuracy and latency predictors to reduce the cost of NAS. We introduce a novel search-space independent NN encoding based on zero-cost proxies that achieves sample-efficient prediction on multiple tasks and NAS search spaces, improving the end-to-end sample efficiency of latency and accuracy predictors by over an order of magnitude in multiple scenarios. For example, our NN encoding enables multi-search-space transfer of latency predictors from NASBench-201 to FBNet (and vice-versa) in under 85 HW measurements, a 400$\times$ improvement in sample efficiency compared to a recent meta-learning approach. Our method also improves the total sample efficiency of accuracy predictors by over an order of magnitude. Finally, we demonstrate the effectiveness of our method for multi-search-space and multi-task accuracy prediction on 28 NAS search spaces and tasks.
☆ ESTISR: Adapting Efficient Scene Text Image Super-resolution for Real-Scenes
While scene text image super-resolution (STISR) has yielded remarkable improvements in accurately recognizing scene text, prior methodologies have placed excessive emphasis on optimizing performance, rather than paying due attention to efficiency - a crucial factor in ensuring deployment of the STISR-STR pipeline. In this work, we propose a novel Efficient Scene Text Image Super-resolution (ESTISR) Network for resource-limited deployment platform. ESTISR's functionality primarily depends on two critical components: a CNN-based feature extractor and an efficient self-attention mechanism used for decoding low-resolution images. We designed a re-parameterized inverted residual block specifically suited for resource-limited circumstances as the feature extractor. Meanwhile, we proposed a novel self-attention mechanism, softmax shrinking, based on a kernel-based approach. This innovative technique offers linear complexity while also naturally incorporating discriminating low-level features into the self-attention structure. Extensive experiments on TextZoom show that ESTISR retains a high image restoration quality and improved STR accuracy of low-resolution images. Furthermore, ESTISR consistently outperforms current methods in terms of actual running time and peak memory consumption, while achieving a better trade-off between performance and efficiency.
☆ Sanity Checks for Saliency Methods Explaining Object Detectors
Saliency methods are frequently used to explain Deep Neural Network-based models. Adebayo et al.'s work on evaluating saliency methods for classification models illustrate certain explanation methods fail the model and data randomization tests. However, on extending the tests for various state of the art object detectors we illustrate that the ability to explain a model is more dependent on the model itself than the explanation method. We perform sanity checks for object detection and define new qualitative criteria to evaluate the saliency explanations, both for object classification and bounding box decisions, using Guided Backpropagation, Integrated Gradients, and their Smoothgrad versions, together with Faster R-CNN, SSD, and EfficientDet-D0, trained on COCO. In addition, the sensitivity of the explanation method to model parameters and data labels varies class-wise motivating to perform the sanity checks for each class. We find that EfficientDet-D0 is the most interpretable method independent of the saliency method, which passes the sanity checks with little problems.
comment: 18 pages, 10 figures, 1st World Conference on eXplainable Artificial Intelligence camera ready
☆ Training Like a Medical Resident: Universal Medical Image Segmentation via Context Prior Learning
A major enduring focus of clinical workflows is disease analytics and diagnosis, leading to medical imaging datasets where the modalities and annotations are strongly tied to specific clinical objectives. To date, building task-specific segmentation models is intuitive yet a restrictive approach, lacking insights gained from widespread imaging cohorts. Inspired by the training of medical residents, we explore universal medical image segmentation, whose goal is to learn from diverse medical imaging sources covering a range of clinical targets, body regions, and image modalities. Following this paradigm, we propose Hermes, a context prior learning approach that addresses the challenges related to the heterogeneity on data, modality, and annotations in the proposed universal paradigm. In a collection of seven diverse datasets, we demonstrate the appealing merits of the universal paradigm over the traditional task-specific training paradigm. By leveraging the synergy among various tasks, Hermes shows superior performance and model scalability. Our in-depth investigation on two additional datasets reveals Hermes' strong capabilities for transfer learning, incremental learning, and generalization to different downstream tasks. The code is available: https://github.com/yhygao/universal-medical-image-segmentation.
☆ Heteroskedastic Geospatial Tracking with Distributed Camera Networks
Visual object tracking has seen significant progress in recent years. However, the vast majority of this work focuses on tracking objects within the image plane of a single camera and ignores the uncertainty associated with predicted object locations. In this work, we focus on the geospatial object tracking problem using data from a distributed camera network. The goal is to predict an object's track in geospatial coordinates along with uncertainty over the object's location while respecting communication constraints that prohibit centralizing raw image data. We present a novel single-object geospatial tracking data set that includes high-accuracy ground truth object locations and video data from a network of four cameras. We present a modeling framework for addressing this task including a novel backbone model and explore how uncertainty calibration and fine-tuning through a differentiable tracker affect performance.
☆ Scale Guided Hypernetwork for Blind Super-Resolution Image Quality Assessment
With the emergence of image super-resolution (SR) algorithm, how to blindly evaluate the quality of super-resolution images has become an urgent task. However, existing blind SR image quality assessment (IQA) metrics merely focus on visual characteristics of super-resolution images, ignoring the available scale information. In this paper, we reveal that the scale factor has a statistically significant impact on subjective quality scores of SR images, indicating that the scale information can be used to guide the task of blind SR IQA. Motivated by this, we propose a scale guided hypernetwork framework that evaluates SR image quality in a scale-adaptive manner. Specifically, the blind SR IQA procedure is divided into three stages, i.e., content perception, evaluation rule generation, and quality prediction. After content perception, a hypernetwork generates the evaluation rule used in quality prediction based on the scale factor of the SR image. We apply the proposed scale guided hypernetwork framework to existing representative blind IQA metrics, and experimental results show that the proposed framework not only boosts the performance of these IQA metrics but also enhances their generalization abilities. Source code will be available at https://github.com/JunFu1995/SGH.
comment: new framework for blind super-resolution image quality assessment
☆ NICE-SLAM with Adaptive Feature Grids
NICE-SLAM is a dense visual SLAM system that combines the advantages of neural implicit representations and hierarchical grid-based scene representation. However, the hierarchical grid features are densely stored, leading to memory explosion problems when adapting the framework to large scenes. In our project, we present sparse NICE-SLAM, a sparse SLAM system incorporating the idea of Voxel Hashing into NICE-SLAM framework. Instead of initializing feature grids in the whole space, voxel features near the surface are adaptively added and optimized. Experiments demonstrated that compared to NICE-SLAM algorithm, our approach takes much less memory and achieves comparable reconstruction quality on the same datasets. Our implementation is available at https://github.com/zhangganlin/NICE-SLAM-with-Adaptive-Feature-Grids.
comment: Course project of 3D Vision at ETH Zurich
☆ Accessible Robot Control in Mixed Reality
A novel method to control the Spot robot of Boston Dynamics by Hololens 2 is proposed. This method is mainly designed for people with physical disabilities, users can control the robot's movement and robot arm without using their hands. The eye gaze tracking and head motion tracking technologies of Hololens 2 are utilized for sending control commands. The movement of the robot would follow the eye gaze and the robot arm would mimic the pose of the user's head. Through our experiment, our method is comparable with the traditional control method by joystick in both time efficiency and user experience. Demo can be found on our project webpage: https://zhangganlin.github.io/Holo-Spot-Page/index.html
comment: Course Project of Mixed Reality at ETH Zurich
☆ GAN-based Deidentification of Drivers' Face Videos: An Assessment of Human Factors Implications in NDS Data
This paper addresses the problem of sharing drivers' face videos for transportation research while adhering to proper ethical guidelines. The paper first gives an overview of the multitude of problems associated with sharing such data and then proposes a framework on how artificial intelligence-based techniques, specifically face swapping, can be used for de-identifying drivers' faces. Through extensive experimentation with an Oak Ridge National Laboratory (ORNL) dataset, we demonstrate the effectiveness of face-swapping algorithms in preserving essential attributes related to human factors research, including eye movements, head movements, and mouth movements. The efficacy of the framework was also tested on various naturalistic driving study data collected at the Virginia Tech Transportation Institute. The results achieved through the proposed techniques were evaluated qualitatively and quantitatively using various metrics. Finally, we discuss possible measures for sharing the de-identified videos with the greater research community.
comment: Accepted in IEEE IV 2023
☆ RSSOD-Bench: A large-scale benchmark dataset for Salient Object Detection in Optical Remote Sensing Imagery
We present the RSSOD-Bench dataset for salient object detection (SOD) in optical remote sensing imagery. While SOD has achieved success in natural scene images with deep learning, research in SOD for remote sensing imagery (RSSOD) is still in its early stages. Existing RSSOD datasets have limitations in terms of scale, and scene categories, which make them misaligned with real-world applications. To address these shortcomings, we construct the RSSOD-Bench dataset, which contains images from four different cities in the USA. The dataset provides annotations for various salient object categories, such as buildings, lakes, rivers, highways, bridges, aircraft, ships, athletic fields, and more. The salient objects in RSSOD-Bench exhibit large-scale variations, cluttered backgrounds, and different seasons. Unlike existing datasets, RSSOD-Bench offers uniform distribution across scene categories. We benchmark 23 different state-of-the-art approaches from both the computer vision and remote sensing communities. Experimental results demonstrate that more research efforts are required for the RSSOD task.
comment: IGARSS 2023, 4 pages
☆ CDLT: A Dataset with Concept Drift and Long-Tailed Distribution for Fine-Grained Visual Categorization
Data is the foundation for the development of computer vision, and the establishment of datasets plays an important role in advancing the techniques of fine-grained visual categorization~(FGVC). In the existing FGVC datasets used in computer vision, it is generally assumed that each collected instance has fixed characteristics and the distribution of different categories is relatively balanced. In contrast, the real world scenario reveals the fact that the characteristics of instances tend to vary with time and exhibit a long-tailed distribution. Hence, the collected datasets may mislead the optimization of the fine-grained classifiers, resulting in unpleasant performance in real applications. Starting from the real-world conditions and to promote the practical progress of fine-grained visual categorization, we present a Concept Drift and Long-Tailed Distribution dataset. Specifically, the dataset is collected by gathering 11195 images of 250 instances in different species for 47 consecutive months in their natural contexts. The collection process involves dozens of crowd workers for photographing and domain experts for labelling. Extensive baseline experiments using the state-of-the-art fine-grained classification models demonstrate the issues of concept drift and long-tailed distribution existed in the dataset, which require the attention of future researches.
☆ Towards Robust Feature Learning with t-vFM Similarity for Continual Learning
Continual learning has been developed using standard supervised contrastive loss from the perspective of feature learning. Due to the data imbalance during the training, there are still challenges in learning better representations. In this work, we suggest using a different similarity metric instead of cosine similarity in supervised contrastive loss in order to learn more robust representations. We validate the our method on one of the image classification datasets Seq-CIFAR-10 and the results outperform recent continual learning baselines.
☆ Multi-CLIP: Contrastive Vision-Language Pre-training for Question Answering tasks in 3D Scenes
Training models to apply common-sense linguistic knowledge and visual concepts from 2D images to 3D scene understanding is a promising direction that researchers have only recently started to explore. However, it still remains understudied whether 2D distilled knowledge can provide useful representations for downstream 3D vision-language tasks such as 3D question answering. In this paper, we propose a novel 3D pre-training Vision-Language method, namely Multi-CLIP, that enables a model to learn language-grounded and transferable 3D scene point cloud representations. We leverage the representational power of the CLIP model by maximizing the agreement between the encoded 3D scene features and the corresponding 2D multi-view image and text embeddings in the CLIP space via a contrastive objective. To validate our approach, we consider the challenging downstream tasks of 3D Visual Question Answering (3D-VQA) and 3D Situated Question Answering (3D-SQA). To this end, we develop novel multi-modal transformer-based architectures and we demonstrate how our pre-training method can benefit their performance. Quantitative and qualitative experimental results show that Multi-CLIP outperforms state-of-the-art works across the downstream tasks of 3D-VQA and 3D-SQA and leads to a well-structured 3D scene feature space.
comment: The first two authors contributed equally
☆ Temporal Dynamic Quantization for Diffusion Models
The diffusion model has gained popularity in vision applications due to its remarkable generative performance and versatility. However, high storage and computation demands, resulting from the model size and iterative generation, hinder its use on mobile devices. Existing quantization techniques struggle to maintain performance even in 8-bit precision due to the diffusion model's unique property of temporal variation in activation. We introduce a novel quantization method that dynamically adjusts the quantization interval based on time step information, significantly improving output quality. Unlike conventional dynamic quantization techniques, our approach has no computational overhead during inference and is compatible with both post-training quantization (PTQ) and quantization-aware training (QAT). Our extensive experiments demonstrate substantial improvements in output quality with the quantized diffusion model across various datasets.
☆ Using Unreliable Pseudo-Labels for Label-Efficient Semantic Segmentation
The crux of label-efficient semantic segmentation is to produce high-quality pseudo-labels to leverage a large amount of unlabeled or weakly labeled data. A common practice is to select the highly confident predictions as the pseudo-ground-truths for each pixel, but it leads to a problem that most pixels may be left unused due to their unreliability. However, we argue that every pixel matters to the model training, even those unreliable and ambiguous pixels. Intuitively, an unreliable prediction may get confused among the top classes, however, it should be confident about the pixel not belonging to the remaining classes. Hence, such a pixel can be convincingly treated as a negative key to those most unlikely categories. Therefore, we develop an effective pipeline to make sufficient use of unlabeled data. Concretely, we separate reliable and unreliable pixels via the entropy of predictions, push each unreliable pixel to a category-wise queue that consists of negative keys, and manage to train the model with all candidate pixels. Considering the training evolution, we adaptively adjust the threshold for the reliable-unreliable partition. Experimental results on various benchmarks and training settings demonstrate the superiority of our approach over the state-of-the-art alternatives.
☆ Cross-CBAM: A Lightweight network for Scene Segmentation
Scene parsing is a great challenge for real-time semantic segmentation. Although traditional semantic segmentation networks have made remarkable leap-forwards in semantic accuracy, the performance of inference speed is unsatisfactory. Meanwhile, this progress is achieved with fairly large networks and powerful computational resources. However, it is difficult to run extremely large models on edge computing devices with limited computing power, which poses a huge challenge to the real-time semantic segmentation tasks. In this paper, we present the Cross-CBAM network, a novel lightweight network for real-time semantic segmentation. Specifically, a Squeeze-and-Excitation Atrous Spatial Pyramid Pooling Module(SE-ASPP) is proposed to get variable field-of-view and multiscale information. And we propose a Cross Convolutional Block Attention Module(CCBAM), in which a cross-multiply operation is employed in the CCBAM module to make high-level semantic information guide low-level detail information. Different from previous work, these works use attention to focus on the desired information in the backbone. CCBAM uses cross-attention for feature fusion in the FPN structure. Extensive experiments on the Cityscapes dataset and Camvid dataset demonstrate the effectiveness of the proposed Cross-CBAM model by achieving a promising trade-off between segmentation accuracy and inference speed. On the Cityscapes test set, we achieve 73.4% mIoU with a speed of 240.9FPS and 77.2% mIoU with a speed of 88.6FPS on NVIDIA GTX 1080Ti.
☆ rPPG-MAE: Self-supervised Pre-training with Masked Autoencoders for Remote Physiological Measurement
Remote photoplethysmography (rPPG) is an important technique for perceiving human vital signs, which has received extensive attention. For a long time, researchers have focused on supervised methods that rely on large amounts of labeled data. These methods are limited by the requirement for large amounts of data and the difficulty of acquiring ground truth physiological signals. To address these issues, several self-supervised methods based on contrastive learning have been proposed. However, they focus on the contrastive learning between samples, which neglect the inherent self-similar prior in physiological signals and seem to have a limited ability to cope with noisy. In this paper, a linear self-supervised reconstruction task was designed for extracting the inherent self-similar prior in physiological signals. Besides, a specific noise-insensitive strategy was explored for reducing the interference of motion and illumination. The proposed framework in this paper, namely rPPG-MAE, demonstrates excellent performance even on the challenging VIPL-HR dataset. We also evaluate the proposed method on two public datasets, namely PURE and UBFC-rPPG. The results show that our method not only outperforms existing self-supervised methods but also exceeds the state-of-the-art (SOTA) supervised methods. One important observation is that the quality of the dataset seems more important than the size in self-supervised pre-training of rPPG. The source code is released at https://github.com/linuxsino/rPPG-MAE.
☆ 3rd Place Solution for PVUW2023 VSS Track: A Large Model for Semantic Segmentation on VSPW CVPR 2023
In this paper, we introduce 3rd place solution for PVUW2023 VSS track. Semantic segmentation is a fundamental task in computer vision with numerous real-world applications. We have explored various image-level visual backbones and segmentation heads to tackle the problem of video semantic segmentation. Through our experimentation, we find that InternImage-H as the backbone and Mask2former as the segmentation head achieves the best performance. In addition, we explore two post-precessing methods: CascadePSP and Segment Anything Model (SAM). Ultimately, our approach obtains 62.60\% and 64.84\% mIoU on the VSPW test set1 and final test set, respectively, securing the third position in the PVUW2023 VSS track.
comment: 3rd Place Solution for CVPR 2023 PVUW VSS Track
☆ EfficientSRFace: An Efficient Network with Super-Resolution Enhancement for Accurate Face Detection
In face detection, low-resolution faces, such as numerous small faces of a human group in a crowded scene, are common in dense face prediction tasks. They usually contain limited visual clues and make small faces less distinguishable from the other small objects, which poses great challenge to accurate face detection. Although deep convolutional neural network has significantly promoted the research on face detection recently, current deep face detectors rarely take into account low-resolution faces and are still vulnerable to the real-world scenarios where massive amount of low-resolution faces exist. Consequently, they usually achieve degraded performance for low-resolution face detection. In order to alleviate this problem, we develop an efficient detector termed EfficientSRFace by introducing a feature-level super-resolution reconstruction network for enhancing the feature representation capability of the model. This module plays an auxiliary role in the training process, and can be removed during the inference without increasing the inference time. Extensive experiments on public benchmarking datasets, such as FDDB and WIDER Face, show that the embedded image super-resolution module can significantly improve the detection accuracy at the cost of a small amount of additional parameters and computational overhead, while helping our model achieve competitive performance compared with the state-of-the-arts methods.
☆ USD: Unknown Sensitive Detector Empowered by Decoupled Objectness and Segment Anything Model
Open World Object Detection (OWOD) is a novel and challenging computer vision task that enables object detection with the ability to detect unknown objects. Existing methods typically estimate the object likelihood with an additional objectness branch, but ignore the conflict in learning objectness and classification boundaries, which oppose each other on the semantic manifold and training objective. To address this issue, we propose a simple yet effective learning strategy, namely Decoupled Objectness Learning (DOL), which divides the learning of these two boundaries into suitable decoder layers. Moreover, detecting unknown objects comprehensively requires a large amount of annotations, but labeling all unknown objects is both difficult and expensive. Therefore, we propose to take advantage of the recent Large Vision Model (LVM), specifically the Segment Anything Model (SAM), to enhance the detection of unknown objects. Nevertheless, the output results of SAM contain noise, including backgrounds and fragments, so we introduce an Auxiliary Supervision Framework (ASF) that uses a pseudo-labeling and a soft-weighting strategies to alleviate the negative impact of noise. Extensive experiments on popular benchmarks, including Pascal VOC and MS COCO, demonstrate the effectiveness of our approach. Our proposed Unknown Sensitive Detector (USD) outperforms the recent state-of-the-art methods in terms of Unknown Recall, achieving significant improvements of 14.3\%, 15.5\%, and 8.9\% on the M-OWODB, and 27.1\%, 29.1\%, and 25.1\% on the S-OWODB.
☆ Revisiting Class Imbalance for End-to-end Semi-Supervised Object Detection CVPR 2023
Semi-supervised object detection (SSOD) has made significant progress with the development of pseudo-label-based end-to-end methods. However, many of these methods face challenges due to class imbalance, which hinders the effectiveness of the pseudo-label generator. Furthermore, in the literature, it has been observed that low-quality pseudo-labels severely limit the performance of SSOD. In this paper, we examine the root causes of low-quality pseudo-labels and present novel learning mechanisms to improve the label generation quality. To cope with high false-negative and low precision rates, we introduce an adaptive thresholding mechanism that helps the proposed network to filter out optimal bounding boxes. We further introduce a Jitter-Bagging module to provide accurate information on localization to help refine the bounding boxes. Additionally, two new losses are introduced using the background and foreground scores predicted by the teacher and student networks to improvise the pseudo-label recall rate. Furthermore, our method applies strict supervision to the teacher network by feeding strong & weak augmented data to generate robust pseudo-labels so that it can detect small and complex objects. Finally, the extensive experiments show that the proposed network outperforms state-of-the-art methods on MS-COCO and Pascal VOC datasets and allows the baseline network to achieve 100% supervised performance with much less (i.e., 20%) labeled data.
comment: Accepted at the Efficient Deep Learning for Computer Vision Workshop, CVPR 2023
☆ MAVD: The First Open Large-Scale Mandarin Audio-Visual Dataset with Depth Information
Audio-visual speech recognition (AVSR) gains increasing attention from researchers as an important part of human-computer interaction. However, the existing available Mandarin audio-visual datasets are limited and lack the depth information. To address this issue, this work establishes the MAVD, a new large-scale Mandarin multimodal corpus comprising 12,484 utterances spoken by 64 native Chinese speakers. To ensure the dataset covers diverse real-world scenarios, a pipeline for cleaning and filtering the raw text material has been developed to create a well-balanced reading material. In particular, the latest data acquisition device of Microsoft, Azure Kinect is used to capture depth information in addition to the traditional audio signals and RGB images during data acquisition. We also provide a baseline experiment, which could be used to evaluate the effectiveness of the dataset. The dataset and code will be released at https://github.com/SpringHuo/MAVD.
☆ MoviePuzzle: Visual Narrative Reasoning through Multimodal Order Learning
We introduce MoviePuzzle, a novel challenge that targets visual narrative reasoning and holistic movie understanding. Despite the notable progress that has been witnessed in the realm of video understanding, most prior works fail to present tasks and models to address holistic video understanding and the innate visual narrative structures existing in long-form videos. To tackle this quandary, we put forth MoviePuzzle task that amplifies the temporal feature learning and structure learning of video models by reshuffling the shot, frame, and clip layers of movie segments in the presence of video-dialogue information. We start by establishing a carefully refined dataset based on MovieNet by dissecting movies into hierarchical layers and randomly permuting the orders. Besides benchmarking the MoviePuzzle with prior arts on movie understanding, we devise a Hierarchical Contrastive Movie Clustering (HCMC) model that considers the underlying structure and visual semantic orders for movie reordering. Specifically, through a pairwise and contrastive learning approach, we train models to predict the correct order of each layer. This equips them with the knack for deciphering the visual narrative structure of movies and handling the disorder lurking in video data. Experiments show that our approach outperforms existing state-of-the-art methods on the \MoviePuzzle benchmark, underscoring its efficacy.
☆ SAM3D: Zero-Shot 3D Object Detection via Segment Anything Model
With the development of large language models, many remarkable linguistic systems like ChatGPT have thrived and achieved astonishing success on many tasks, showing the incredible power of foundation models. In the spirit of unleashing the capability of foundation models on vision tasks, the Segment Anything Model (SAM), a vision foundation model for image segmentation, has been proposed recently and presents strong zero-shot ability on many downstream 2D tasks. However, whether SAM can be adapted to 3D vision tasks has yet to be explored, especially 3D object detection. With this inspiration, we explore adapting the zero-shot ability of SAM to 3D object detection in this paper. We propose a SAM-powered BEV processing pipeline to detect objects and get promising results on the large-scale Waymo open dataset. As an early attempt, our method takes a step toward 3D object detection with vision foundation models and presents the opportunity to unleash their power on 3D vision tasks. The code is released at https://github.com/DYZhang09/SAM3D.
comment: Technical Report. The code is released at https://github.com/DYZhang09/SAM3D
♻ ☆ The Canadian Cropland Dataset: A New Land Cover Dataset for Multitemporal Deep Learning Classification in Agriculture
Monitoring land cover using remote sensing is vital for studying environmental changes and ensuring global food security through crop yield forecasting. Specifically, multitemporal remote sensing imagery provides relevant information about the dynamics of a scene, which has proven to lead to better land cover classification results. Nevertheless, few studies have benefited from high spatial and temporal resolution data due to the difficulty of accessing reliable, fine-grained and high-quality annotated samples to support their hypotheses. Therefore, we introduce a temporal patch-based dataset of Canadian croplands, enriched with labels retrieved from the Canadian Annual Crop Inventory. The dataset contains 78,536 manually verified high-resolution (10 m/pixel, 640 x 640 m) geo-referenced images from 10 crop classes collected over four crop production years (2017-2020) and five months (June-October). Each instance contains 12 spectral bands, an RGB image, and additional vegetation index bands. Individually, each category contains at least 4,800 images. Moreover, as a benchmark, we provide models and source code that allow a user to predict the crop class using a single image (ResNet, DenseNet, EfficientNet) or a sequence of images (LRCN, 3D-CNN) from the same location. In perspective, we expect this evolving dataset to propel the creation of robust agro-environmental models that can accelerate the comprehension of complex agricultural regions by providing accurate and continuous monitoring of land cover.
comment: 24 pages, 5 figures, dataset descriptor
♻ ☆ Refining Generative Process with Discriminator Guidance in Score-based Diffusion Models ICML23
The proposed method, Discriminator Guidance, aims to improve sample generation of pre-trained diffusion models. The approach introduces a discriminator that gives explicit supervision to a denoising sample path whether it is realistic or not. Unlike GANs, our approach does not require joint training of score and discriminator networks. Instead, we train the discriminator after score training, making discriminator training stable and fast to converge. In sample generation, we add an auxiliary term to the pre-trained score to deceive the discriminator. This term corrects the model score to the data score at the optimal discriminator, which implies that the discriminator helps better score estimation in a complementary way. Using our algorithm, we achive state-of-the-art results on ImageNet 256x256 with FID 1.83 and recall 0.64, similar to the validation data's FID (1.68) and recall (0.66). We release the code at https://github.com/alsdudrla10/DG.
comment: International Conference on Machine Learning (ICML23)
♻ ☆ DExT: Detector Explanation Toolkit
State-of-the-art object detectors are treated as black boxes due to their highly non-linear internal computations. Even with unprecedented advancements in detector performance, the inability to explain how their outputs are generated limits their use in safety-critical applications. Previous work fails to produce explanations for both bounding box and classification decisions, and generally make individual explanations for various detectors. In this paper, we propose an open-source Detector Explanation Toolkit (DExT) which implements the proposed approach to generate a holistic explanation for all detector decisions using certain gradient-based explanation methods. We suggests various multi-object visualization methods to merge the explanations of multiple objects detected in an image as well as the corresponding detections in a single image. The quantitative evaluation show that the Single Shot MultiBox Detector (SSD) is more faithfully explained compared to other detectors regardless of the explanation methods. Both quantitative and human-centric evaluations identify that SmoothGrad with Guided Backpropagation (GBP) provides more trustworthy explanations among selected methods across all detectors. We expect that DExT will motivate practitioners to evaluate object detectors from the interpretability perspective by explaining both bounding box and classification decisions.
comment: 24 pages, with appendix. 1st World Conference on eXplainable Artificial Intelligence camera ready
♻ ☆ PaintSeg: Training-free Segmentation via Painting
The paper introduces PaintSeg, a new unsupervised method for segmenting objects without any training. We propose an adversarial masked contrastive painting (AMCP) process, which creates a contrast between the original image and a painted image in which a masked area is painted using off-the-shelf generative models. During the painting process, inpainting and outpainting are alternated, with the former masking the foreground and filling in the background, and the latter masking the background while recovering the missing part of the foreground object. Inpainting and outpainting, also referred to as I-step and O-step, allow our method to gradually advance the target segmentation mask toward the ground truth without supervision or training. PaintSeg can be configured to work with a variety of prompts, e.g. coarse masks, boxes, scribbles, and points. Our experimental results demonstrate that PaintSeg outperforms existing approaches in coarse mask-prompt, box-prompt, and point-prompt segmentation tasks, providing a training-free solution suitable for unsupervised segmentation.
♻ ☆ Intriguing Properties of Text-guided Diffusion Models
Text-guided diffusion models (TDMs) are widely applied but can fail unexpectedly. Common failures include: (i) natural-looking text prompts generating images with the wrong content, or (ii) different random samples of the latent variables that generate vastly different, and even unrelated, outputs despite being conditioned on the same text prompt. In this work, we aim to study and understand the failure modes of TDMs in more detail. To achieve this, we propose SAGE, an adversarial attack on TDMs that uses image classifiers as surrogate loss functions, to search over the discrete prompt space and the high-dimensional latent space of TDMs to automatically discover unexpected behaviors and failure cases in the image generation. We make several technical contributions to ensure that SAGE finds failure cases of the diffusion model, rather than the classifier, and verify this in a human study. Our study reveals four intriguing properties of TDMs that have not been systematically studied before: (1) We find a variety of natural text prompts producing images that fail to capture the semantics of input texts. We categorize these failures into ten distinct types based on the underlying causes. (2) We find samples in the latent space (which are not outliers) that lead to distorted images independent of the text prompt, suggesting that parts of the latent space are not well-structured. (3) We also find latent samples that lead to natural-looking images which are unrelated to the text prompt, implying a potential misalignment between the latent and prompt spaces. (4) By appending a single adversarial token embedding to an input prompt we can generate a variety of specified target objects, while only minimally affecting the CLIP score. This demonstrates the fragility of language representations and raises potential safety concerns.
comment: Code will be available at: https://github.com/qihao067/SAGE
♻ ☆ Automatic 3D Registration of Dental CBCT and Face Scan Data using 2D Projection images
This paper presents a fully automatic registration method of dental cone-beam computed tomography (CBCT) and face scan data. It can be used for a digital platform of 3D jaw-teeth-face models in a variety of applications, including 3D digital treatment planning and orthognathic surgery. Difficulties in accurately merging facial scans and CBCT images are due to the different image acquisition methods and limited area of correspondence between the two facial surfaces. In addition, it is difficult to use machine learning techniques because they use face-related 3D medical data with radiation exposure, which are difficult to obtain for training. The proposed method addresses these problems by reusing an existing machine-learning-based 2D landmark detection algorithm in an open-source library and developing a novel mathematical algorithm that identifies paired 3D landmarks from knowledge of the corresponding 2D landmarks. A main contribution of this study is that the proposed method does not require annotated training data of facial landmarks because it uses a pre-trained facial landmark detection algorithm that is known to be robust and generalized to various 2D face image models. Note that this reduces a 3D landmark detection problem to a 2D problem of identifying the corresponding landmarks on two 2D projection images generated from two different projection angles. Here, the 3D landmarks for registration were selected from the sub-surfaces with the least geometric change under the CBCT and face scan environments. For the final fine-tuning of the registration, the Iterative Closest Point method was applied, which utilizes geometrical information around the 3D landmarks. The experimental results show that the proposed method achieved an averaged surface distance error of 0.74 mm for three pairs of CBCT and face scan datasets.
comment: 8 pages, 6 figures, 2 tables
♻ ☆ On Investigating the Conservative Property of Score-Based Generative Models ICML 2023
Existing Score-Based Models (SBMs) can be categorized into constrained SBMs (CSBMs) or unconstrained SBMs (USBMs) according to their parameterization approaches. CSBMs model probability density functions as Boltzmann distributions, and assign their predictions as the negative gradients of some scalar-valued energy functions. On the other hand, USBMs employ flexible architectures capable of directly estimating scores without the need to explicitly model energy functions. In this paper, we demonstrate that the architectural constraints of CSBMs may limit their modeling ability. In addition, we show that USBMs' inability to preserve the property of conservativeness may lead to degraded performance in practice. To address the above issues, we propose Quasi-Conservative Score-Based Models (QCSBMs) for keeping the advantages of both CSBMs and USBMs. Our theoretical derivations demonstrate that the training objective of QCSBMs can be efficiently integrated into the training processes by leveraging the Hutchinson's trace estimator. In addition, our experimental results on the CIFAR-10, CIFAR-100, ImageNet, and SVHN datasets validate the effectiveness of QCSBMs. Finally, we justify the advantage of QCSBMs using an example of a one-layered autoencoder.
comment: ICML 2023
♻ ☆ Learning with Explicit Shape Priors for Medical Image Segmentation
Medical image segmentation is a fundamental task for medical image analysis and surgical planning. In recent years, UNet-based networks have prevailed in the field of medical image segmentation. However, convolution-neural networks (CNNs) suffer from limited receptive fields, which fail to model the long-range dependency of organs or tumors. Besides, these models are heavily dependent on the training of the final segmentation head. And existing methods can not well address these two limitations at the same time. Hence, in our work, we proposed a novel shape prior module (SPM), which can explicitly introduce shape priors to promote the segmentation performance of UNet-based models. The explicit shape priors consist of global and local shape priors. The former with coarse shape representations provides networks with capabilities to model global contexts. The latter with finer shape information serves as additional guidance to boost the segmentation performance, which relieves the heavy dependence on the learnable prototype in the segmentation head. To evaluate the effectiveness of SPM, we conduct experiments on three challenging public datasets. And our proposed model achieves state-of-the-art performance. Furthermore, SPM shows an outstanding generalization ability on classic CNNs and recent Transformer-based backbones, which can serve as a plug-and-play structure for the segmentation task of different datasets. Source codes are available at https://github.com/AlexYouXin/Explicit-Shape-Priors
comment: 10 pages, 10 figures
♻ ☆ Alignment with human representations supports robust few-shot learning
Should we care whether AI systems have representations of the world that are similar to those of humans? We provide an information-theoretic analysis that suggests that there should be a U-shaped relationship between the degree of representational alignment with humans and performance on few-shot learning tasks. We confirm this prediction empirically, finding such a relationship in an analysis of the performance of 491 computer vision models. We also show that highly-aligned models are more robust to both adversarial attacks and domain shifts. Our results suggest that human-alignment is often a sufficient, but not necessary, condition for models to make effective use of limited data, be robust, and generalize well.
♻ ☆ VoxDet: Voxel Learning for Novel Instance Detection
Detecting unseen instances based on multi-view templates is a challenging problem due to its open-world nature. Traditional methodologies, which primarily rely on 2D representations and matching techniques, are often inadequate in handling pose variations and occlusions. To solve this, we introduce VoxDet, a pioneer 3D geometry-aware framework that fully utilizes the strong 3D voxel representation and reliable voxel matching mechanism. VoxDet first ingeniously proposes template voxel aggregation (TVA) module, effectively transforming multi-view 2D images into 3D voxel features. By leveraging associated camera poses, these features are aggregated into a compact 3D template voxel. In novel instance detection, this voxel representation demonstrates heightened resilience to occlusion and pose variations. We also discover that a 3D reconstruction objective helps to pre-train the 2D-3D mapping in TVA. Second, to quickly align with the template voxel, VoxDet incorporates a Query Voxel Matching (QVM) module. The 2D queries are first converted into their voxel representation with the learned 2D-3D mapping. We find that since the 3D voxel representations encode the geometry, we can first estimate the relative rotation and then compare the aligned voxels, leading to improved accuracy and efficiency. Exhaustive experiments are conducted on the demanding LineMod-Occlusion, YCB-video, and the newly built RoboTools benchmarks, where VoxDet outperforms various 2D baselines remarkably with 20% higher recall and faster speed. To the best of our knowledge, VoxDet is the first to incorporate implicit 3D knowledge for 2D detection tasks.
comment: 17 pages, 10 figures
♻ ☆ Direct Learning-Based Deep Spiking Neural Networks: A Review
The spiking neural network (SNN), as a promising brain-inspired computational model with binary spike information transmission mechanism, rich spatially-temporal dynamics, and event-driven characteristics, has received extensive attention. However, its intricately discontinuous spike mechanism brings difficulty to the optimization of the deep SNN. Since the surrogate gradient method can greatly mitigate the optimization difficulty and shows great potential in directly training deep SNNs, a variety of direct learning-based deep SNN works have been proposed and achieved satisfying progress in recent years. In this paper, we present a comprehensive survey of these direct learning-based deep SNN works, mainly categorized into accuracy improvement methods, efficiency improvement methods, and temporal dynamics utilization methods. In addition, we also divide these categorizations into finer granularities further to better organize and introduce them. Finally, the challenges and trends that may be faced in future research are prospected.
comment: Accepted by Frontiers in Neuroscience. If your relevant work is omitted, feel free to email me at yfguo@pku.edu.cn
♻ ☆ EnfoMax: Domain Entropy and Mutual Information Maximization for Domain Generalized Face Anti-spoofing
The face anti-spoofing (FAS) method performs well under intra-domain setups. However, its cross-domain performance is unsatisfactory. As a result, the domain generalization (DG) method has gained more attention in FAS. Existing methods treat FAS as a simple binary classification task and propose a heuristic training objective to learn domain-invariant features. However, there is no theoretical explanation of what a domain-invariant feature is. Additionally, the lack of theoretical support makes domain generalization techniques such as adversarial training lack training stability. To address these issues, this paper proposes the EnfoMax framework, which uses information theory to analyze cross-domain FAS tasks. This framework provides theoretical guarantees and optimization objectives for domain-generalized FAS tasks. EnfoMax maximizes the domain entropy and mutual information of live samples in source domains without using adversarial learning. Experimental results demonstrate that our approach performs well on extensive public datasets and outperforms state-of-the-art methods.
♻ ☆ MonoNeRF: Learning Generalizable NeRFs from Monocular Videos without Camera Pose ICML 2023
We propose a generalizable neural radiance fields - MonoNeRF, that can be trained on large-scale monocular videos of moving in static scenes without any ground-truth annotations of depth and camera poses. MonoNeRF follows an Autoencoder-based architecture, where the encoder estimates the monocular depth and the camera pose, and the decoder constructs a Multiplane NeRF representation based on the depth encoder feature, and renders the input frames with the estimated camera. The learning is supervised by the reconstruction error. Once the model is learned, it can be applied to multiple applications including depth estimation, camera pose estimation, and single-image novel view synthesis. More qualitative results are available at: https://oasisyang.github.io/mononerf .
comment: ICML 2023 camera ready version. Project page: https://oasisyang.github.io/mononerf
♻ ☆ SOSR: Source-Free Image Super-Resolution with Wavelet Augmentation Transformer
Real-world images taken by different cameras with different degradation kernels often result in a cross-device domain gap in image super-resolution. A prevalent attempt to this issue is unsupervised domain adaptation (UDA) that needs to access source data. Considering privacy policies or transmission restrictions of data in many practical applications, we propose a SOurce-free image Super-Resolution framework (SOSR) to address this issue, i.e., adapt a model pre-trained on labeled source data to a target domain with only unlabeled target data. SOSR leverages the source model to generate refined pseudo-labels for teacher-student learning. To better utilize the pseudo-labels, this paper proposes a novel wavelet-based augmentation method, named Wavelet Augmentation Transformer (WAT), which can be flexibly incorporated with existing networks, to implicitly produce useful augmented data. WAT learns low-frequency information of varying levels across diverse samples, which is aggregated efficiently via deformable attention. Furthermore, an uncertainty-aware self-training mechanism is proposed to improve the accuracy of pseudo-labels, with inaccurate predictions being rectified by uncertainty estimation. To acquire better SR results and avoid overfitting pseudo-labels, several regularization losses are proposed to constrain the frequency information between target LR and SR images. Experiments show that without accessing source data, SOSR achieves superior results to the state-of-the-art UDA methods.
comment: 15 pages, 9 figures, 10 tables
♻ ☆ Incremental Neural Implicit Representation with Uncertainty-Filtered Knowledge Distillation
Recent neural implicit representations (NIRs) have achieved great success in the tasks of 3D reconstruction and novel view synthesis. However, they suffer from the catastrophic forgetting problem when continuously learning from streaming data without revisiting the previously seen data. This limitation prohibits the application of existing NIRs to scenarios where images come in sequentially. In view of this, we explore the task of incremental learning for NIRs in this work. We design a student-teacher framework to mitigate the catastrophic forgetting problem. Specifically, we iterate the process of using the student as the teacher at the end of each time step and let the teacher guide the training of the student in the next step. As a result, the student network is able to learn new information from the streaming data and retain old knowledge from the teacher network simultaneously. Although intuitive, naively applying the student-teacher pipeline does not work well in our task. Not all information from the teacher network is helpful since it is only trained with the old data. To alleviate this problem, we further introduce a random inquirer and an uncertainty-based filter to filter useful information. Our proposed method is general and thus can be adapted to different implicit representations such as neural radiance field (NeRF) and neural SDF. Extensive experimental results for both 3D reconstruction and novel view synthesis demonstrate the effectiveness of our approach compared to different baselines.
Information Retrieval 13
☆ I^3 Retriever: Incorporating Implicit Interaction in Pre-trained Language Models for Passage Retrieval
Passage retrieval is a fundamental task in many information systems, such as web search and question answering, where both efficiency and effectiveness are critical concerns. In recent years, neural retrievers based on pre-trained language models (PLM), such as dual-encoders, have achieved huge success. Yet, studies have found that the performance of dual-encoders are often limited due to the neglecting of the interaction information between queries and candidate passages. Therefore, various interaction paradigms have been proposed to improve the performance of vanilla dual-encoders. Particularly, recent state-of-the-art methods often introduce late-interaction during the model inference process. However, such late-interaction based methods usually bring extensive computation and storage cost on large corpus. Despite their effectiveness, the concern of efficiency and space footprint is still an important factor that limits the application of interaction-based neural retrieval models. To tackle this issue, we incorporate implicit interaction into dual-encoders, and propose I^3 retriever. In particular, our implicit interaction paradigm leverages generated pseudo-queries to simulate query-passage interaction, which jointly optimizes with query and passage encoders in an end-to-end manner. It can be fully pre-computed and cached, and its inference process only involves simple dot product operation of the query vector and passage vector, which makes it as efficient as the vanilla dual encoders. We conduct comprehensive experiments on MSMARCO and TREC2019 Deep Learning Datasets, demonstrating the I^3 retriever's superiority in terms of both effectiveness and efficiency. Moreover, the proposed implicit interaction is compatible with special pre-training and knowledge distillation for passage retrieval, which brings a new state-of-the-art performance.
comment: 10 pages
☆ bgGLUE: A Bulgarian General Language Understanding Evaluation Benchmark ACL 2023
We present bgGLUE (Bulgarian General Language Understanding Evaluation), a benchmark for evaluating language models on Natural Language Understanding (NLU) tasks in Bulgarian. Our benchmark includes NLU tasks targeting a variety of NLP problems (e.g., natural language inference, fact-checking, named entity recognition, sentiment analysis, question answering, etc.) and machine learning tasks (sequence labeling, document-level classification, and regression). We run the first systematic evaluation of pre-trained language models for Bulgarian, comparing and contrasting results across the nine tasks in the benchmark. The evaluation results show strong performance on sequence labeling tasks, but there is a lot of room for improvement for tasks that require more complex reasoning. We make bgGLUE publicly available together with the fine-tuning and the evaluation code, as well as a public leaderboard at https://bgglue.github.io/, and we hope that it will enable further advancements in developing NLU models for Bulgarian.
comment: Accepted to ACL 2023 (Main Conference)
☆ Graph Transformer for Recommendation SIGIR'2023
This paper presents a novel approach to representation learning in recommender systems by integrating generative self-supervised learning with graph transformer architecture. We highlight the importance of high-quality data augmentation with relevant self-supervised pretext tasks for improving performance. Towards this end, we propose a new approach that automates the self-supervision augmentation process through a rationale-aware generative SSL that distills informative user-item interaction patterns. The proposed recommender with Graph TransFormer (GFormer) that offers parameterized collaborative rationale discovery for selective augmentation while preserving global-aware user-item relationships. In GFormer, we allow the rationale-aware SSL to inspire graph collaborative filtering with task-adaptive invariant rationalization in graph transformer. The experimental results reveal that our GFormer has the capability to consistently improve the performance over baselines on different datasets. Several in-depth experiments further investigate the invariant rationale-aware augmentation from various aspects. The source code for this work is publicly available at: https://github.com/HKUDS/GFormer.
comment: Accepted by SIGIR'2023
☆ Large Language Model Augmented Narrative Driven Recommendations
Narrative-driven recommendation (NDR) presents an information access problem where users solicit recommendations with verbose descriptions of their preferences and context, for example, travelers soliciting recommendations for points of interest while describing their likes/dislikes and travel circumstances. These requests are increasingly important with the rise of natural language-based conversational interfaces for search and recommendation systems. However, NDR lacks abundant training data for models, and current platforms commonly do not support these requests. Fortunately, classical user-item interaction datasets contain rich textual data, e.g., reviews, which often describe user preferences and context - this may be used to bootstrap training for NDR models. In this work, we explore using large language models (LLMs) for data augmentation to train NDR models. We use LLMs for authoring synthetic narrative queries from user-item interactions with few-shot prompting and train retrieval models for NDR on synthetic queries and user-item interaction data. Our experiments demonstrate that this is an effective strategy for training small-parameter retrieval models that outperform other retrieval and LLM baselines for narrative-driven recommendation.
comment: Pre-print
☆ Generative Flow Network for Listwise Recommendation
Personalized recommender systems fulfill the daily demands of customers and boost online businesses. The goal is to learn a policy that can generate a list of items that matches the user's demand or interest. While most existing methods learn a pointwise scoring model that predicts the ranking score of each individual item, recent research shows that the listwise approach can further improve the recommendation quality by modeling the intra-list correlations of items that are exposed together. This has motivated the recent list reranking and generative recommendation approaches that optimize the overall utility of the entire list. However, it is challenging to explore the combinatorial space of list actions and existing methods that use cross-entropy loss may suffer from low diversity issues. In this work, we aim to learn a policy that can generate sufficiently diverse item lists for users while maintaining high recommendation quality. The proposed solution, GFN4Rec, is a generative method that takes the insight of the flow network to ensure the alignment between list generation probability and its reward. The key advantages of our solution are the log scale reward matching loss that intrinsically improves the generation diversity and the autoregressive item selection model that captures the item mutual influences while capturing future reward of the list. As validation of our method's effectiveness and its superior diversity during active exploration, we conduct experiments on simulated online environments as well as an offline evaluation framework for two real-world datasets.
comment: 11 pages, 5 figures, 9 tables
☆ ATEM: A Topic Evolution Model for the Detection of Emerging Topics in Scientific Archives
This paper presents ATEM, a novel framework for studying topic evolution in scientific archives. ATEM is based on dynamic topic modeling and dynamic graph embedding techniques that explore the dynamics of content and citations of documents within a scientific corpus. ATEM explores a new notion of contextual emergence for the discovery of emerging interdisciplinary research topics based on the dynamics of citation links in topic clusters. Our experiments show that ATEM can efficiently detect emerging cross-disciplinary topics within the DBLP archive of over five million computer science articles.
♻ ☆ Towards Disentangling Relevance and Bias in Unbiased Learning to Rank KDD
Unbiased learning to rank (ULTR) studies the problem of mitigating various biases from implicit user feedback data such as clicks, and has been receiving considerable attention recently. A popular ULTR approach for real-world applications uses a two-tower architecture, where click modeling is factorized into a relevance tower with regular input features, and a bias tower with bias-relevant inputs such as the position of a document. A successful factorization will allow the relevance tower to be exempt from biases. In this work, we identify a critical issue that existing ULTR methods ignored - the bias tower can be confounded with the relevance tower via the underlying true relevance. In particular, the positions were determined by the logging policy, i.e., the previous production model, which would possess relevance information. We give both theoretical analysis and empirical results to show the negative effects on relevance tower due to such a correlation. We then propose three methods to mitigate the negative confounding effects by better disentangling relevance and bias. Empirical results on both controlled public datasets and a large-scale industry dataset show the effectiveness of the proposed approaches.
comment: Proceedings of the 29th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining
♻ ☆ ANTM: An Aligned Neural Topic Model for Exploring Evolving Topics
This paper presents an algorithmic family of dynamic topic models called Aligned Neural Topic Models (ANTM), which combine novel data mining algorithms to provide a modular framework for discovering evolving topics. ANTM maintains the temporal continuity of evolving topics by extracting time-aware features from documents using advanced pre-trained Large Language Models (LLMs) and employing an overlapping sliding window algorithm for sequential document clustering. This overlapping sliding window algorithm identifies a different number of topics within each time frame and aligns semantically similar document clusters across time periods. This process captures emerging and fading trends across different periods and allows for a more interpretable representation of evolving topics. Experiments on four distinct datasets show that ANTM outperforms probabilistic dynamic topic models in terms of topic coherence and diversity metrics. Moreover, it improves the scalability and flexibility of dynamic topic models by being accessible and adaptable to different types of algorithms. Additionally, a Python package is developed for researchers and scientists who wish to study the trends and evolving patterns of topics in large-scale textual data.
♻ ☆ UnifieR: A Unified Retriever for Large-Scale Retrieval KDD
Large-scale retrieval is to recall relevant documents from a huge collection given a query. It relies on representation learning to embed documents and queries into a common semantic encoding space. According to the encoding space, recent retrieval methods based on pre-trained language models (PLM) can be coarsely categorized into either dense-vector or lexicon-based paradigms. These two paradigms unveil the PLMs' representation capability in different granularities, i.e., global sequence-level compression and local word-level contexts, respectively. Inspired by their complementary global-local contextualization and distinct representing views, we propose a new learning framework, UnifieR which unifies dense-vector and lexicon-based retrieval in one model with a dual-representing capability. Experiments on passage retrieval benchmarks verify its effectiveness in both paradigms. A uni-retrieval scheme is further presented with even better retrieval quality. We lastly evaluate the model on BEIR benchmark to verify its transferability.
comment: To appear at KDD ADS 2023
♻ ☆ Continuous Input Embedding Size Search For Recommender Systems SIGIR'23
Latent factor models are the most popular backbones for today's recommender systems owing to their prominent performance. Latent factor models represent users and items as real-valued embedding vectors for pairwise similarity computation, and all embeddings are traditionally restricted to a uniform size that is relatively large (e.g., 256-dimensional). With the exponentially expanding user base and item catalog in contemporary e-commerce, this design is admittedly becoming memory-inefficient. To facilitate lightweight recommendation, reinforcement learning (RL) has recently opened up opportunities for identifying varying embedding sizes for different users/items. However, challenged by search efficiency and learning an optimal RL policy, existing RL-based methods are restricted to highly discrete, predefined embedding size choices. This leads to a largely overlooked potential of introducing finer granularity into embedding sizes to obtain better recommendation effectiveness under a given memory budget. In this paper, we propose continuous input embedding size search (CIESS), a novel RL-based method that operates on a continuous search space with arbitrary embedding sizes to choose from. In CIESS, we further present an innovative random walk-based exploration strategy to allow the RL policy to efficiently explore more candidate embedding sizes and converge to a better decision. CIESS is also model-agnostic and hence generalizable to a variety of latent factor RSs, whilst experiments on two real-world datasets have shown state-of-the-art performance of CIESS under different memory budgets when paired with three popular recommendation models.
comment: To appear in SIGIR'23
♻ ☆ LexMAE: Lexicon-Bottlenecked Pretraining for Large-Scale Retrieval ICLR 2023
In large-scale retrieval, the lexicon-weighting paradigm, learning weighted sparse representations in vocabulary space, has shown promising results with high quality and low latency. Despite it deeply exploiting the lexicon-representing capability of pre-trained language models, a crucial gap remains between language modeling and lexicon-weighting retrieval -- the former preferring certain or low-entropy words whereas the latter favoring pivot or high-entropy words -- becoming the main barrier to lexicon-weighting performance for large-scale retrieval. To bridge this gap, we propose a brand-new pre-training framework, lexicon-bottlenecked masked autoencoder (LexMAE), to learn importance-aware lexicon representations. Essentially, we present a lexicon-bottlenecked module between a normal language modeling encoder and a weakened decoder, where a continuous bag-of-words bottleneck is constructed to learn a lexicon-importance distribution in an unsupervised fashion. The pre-trained LexMAE is readily transferred to the lexicon-weighting retrieval via fine-tuning. On the ad-hoc retrieval benchmark, MS-Marco, it achieves 42.6% MRR@10 with 45.8 QPS for the passage dataset and 44.4% MRR@100 with 134.8 QPS for the document dataset, by a CPU machine. And LexMAE shows state-of-the-art zero-shot transfer capability on BEIR benchmark with 12 datasets.
comment: Appeared at ICLR 2023
♻ ☆ Tenrec: A Large-scale Multipurpose Benchmark Dataset for Recommender Systems
Existing benchmark datasets for recommender systems (RS) either are created at a small scale or involve very limited forms of user feedback. RS models evaluated on such datasets often lack practical values for large-scale real-world applications. In this paper, we describe Tenrec, a novel and publicly available data collection for RS that records various user feedback from four different recommendation scenarios. To be specific, Tenrec has the following five characteristics: (1) it is large-scale, containing around 5 million users and 140 million interactions; (2) it has not only positive user feedback, but also true negative feedback (vs. one-class recommendation); (3) it contains overlapped users and items across four different scenarios; (4) it contains various types of user positive feedback, in forms of clicks, likes, shares, and follows, etc; (5) it contains additional features beyond the user IDs and item IDs. We verify Tenrec on ten diverse recommendation tasks by running several classical baseline models per task. Tenrec has the potential to become a useful benchmark dataset for a majority of popular recommendation tasks.
♻ ☆ A First Look at LLM-Powered Generative News Recommendation
Personalized news recommendation systems have become essential tools for users to navigate the vast amount of online news content, yet existing news recommenders face significant challenges such as the cold-start problem, user profile modeling, and news content understanding. Previous works have typically followed an inflexible routine to address a particular challenge through model design, but are limited in their ability to understand news content and capture user interests. In this paper, we introduce GENRE, an LLM-powered generative news recommendation framework, which leverages pretrained semantic knowledge from large language models to enrich news data. Our aim is to provide a flexible and unified solution for news recommendation by moving from model design to prompt design. We showcase the use of GENRE for personalized news generation, user profiling, and news summarization. Extensive experiments with various popular recommendation models demonstrate the effectiveness of GENRE. We will publish our code and data for other researchers to reproduce our work.
comment: 11 pages
Computation and Language 21
☆ Generating High-Quality Emotion Arcs For Low-Resource Languages Using Emotion Lexicons
Automatically generated emotion arcs -- that capture how an individual or a population feels over time -- are widely used in industry and research. However, there is little work on evaluating the generated arcs in English (where the emotion resources are available) and no work on generating or evaluating emotion arcs for low-resource languages. Work on generating emotion arcs in low-resource languages such as those indigenous to Africa, the Americas, and Australia is stymied by the lack of emotion-labeled resources and large language models for those languages. Work on evaluating emotion arcs (for any language) is scarce because of the difficulty of establishing the true (gold) emotion arc. Our work, for the first time, systematically and quantitatively evaluates automatically generated emotion arcs. We also compare two common ways of generating emotion arcs: Machine-Learning (ML) models and Lexicon-Only (LexO) methods. By running experiments on 42 diverse datasets in 9 languages, we show that despite being markedly poor at instance level emotion classification, LexO methods are highly accurate at generating emotion arcs when aggregating information from hundreds of instances. (Predicted arcs have correlations ranging from 0.94 to 0.99 with the gold arcs for various emotions.) We also show that for languages with no emotion lexicons, automatic translations of English emotion lexicons can be used to generate high-quality emotion arcs -- correlations above 0.9 with the gold emotion arcs in all six indigenous African languages explored. This opens up avenues for work on emotions in numerous languages from around the world; crucial not only for commerce, public policy, and health research in service of speakers of those languages, but also to draw meaningful conclusions in emotion-pertinent research using information from around the world (thereby avoiding a western-centric bias in research).
comment: 32 pages, 16 figures. arXiv admin note: substantial text overlap with arXiv:2210.07381
☆ SpeechGen: Unlocking the Generative Power of Speech Language Models with Prompts
Large language models (LLMs) have gained considerable attention for Artificial Intelligence Generated Content (AIGC), particularly with the emergence of ChatGPT. However, the direct adaptation of continuous speech to LLMs that process discrete tokens remains an unsolved challenge, hindering the application of LLMs for speech generation. The advanced speech LMs are in the corner, as that speech signals encapsulate a wealth of information, including speaker and emotion, beyond textual data alone. Prompt tuning has demonstrated notable gains in parameter efficiency and competitive performance on some speech classification tasks. However, the extent to which prompts can effectively elicit generation tasks from speech LMs remains an open question. In this paper, we present pioneering research that explores the application of prompt tuning to stimulate speech LMs for various generation tasks, within a unified framework called SpeechGen, with around 10M trainable parameters. The proposed unified framework holds great promise for efficiency and effectiveness, particularly with the imminent arrival of advanced speech LMs, which will significantly enhance the capabilities of the framework. The code and demos of SpeechGen will be available on the project website: \url{https://ga642381.github.io/SpeechPrompt/speechgen}
comment: Work in progress. The first three authors contributed equally
☆ Question-Context Alignment and Answer-Context Dependencies for Effective Answer Sentence Selection INTERSPEECH 2023
Answer sentence selection (AS2) in open-domain question answering finds answer for a question by ranking candidate sentences extracted from web documents. Recent work exploits answer context, i.e., sentences around a candidate, by incorporating them as additional input string to the Transformer models to improve the correctness scoring. In this paper, we propose to improve the candidate scoring by explicitly incorporating the dependencies between question-context and answer-context into the final representation of a candidate. Specifically, we use Optimal Transport to compute the question-based dependencies among sentences in the passage where the answer is extracted from. We then represent these dependencies as edges in a graph and use Graph Convolutional Network to derive the representation of a candidate, a node in the graph. Our proposed model achieves significant improvements on popular AS2 benchmarks, i.e., WikiQA and WDRASS, obtaining new state-of-the-art on all benchmarks.
comment: final copy for INTERSPEECH 2023
☆ LDEB -- Label Digitization with Emotion Binarization and Machine Learning for Emotion Recognition in Conversational Dialogues
Emotion recognition in conversations (ERC) is vital to the advancements of conversational AI and its applications. Therefore, the development of an automated ERC model using the concepts of machine learning (ML) would be beneficial. However, the conversational dialogues present a unique problem where each dialogue depicts nested emotions that entangle the association between the emotional feature descriptors and emotion type (or label). This entanglement that can be multiplied with the presence of data paucity is an obstacle for a ML model. To overcome this problem, we proposed a novel approach called Label Digitization with Emotion Binarization (LDEB) that disentangles the twists by utilizing the text normalization and 7-bit digital encoding techniques and constructs a meaningful feature space for a ML model to be trained. We also utilized the publicly available dataset called the FETA-DailyDialog dataset for feature learning and developed a hierarchical ERC model using random forest (RF) and artificial neural network (ANN) classifiers. Simulations showed that the ANN-based ERC model was able to predict emotion with the best accuracy and precision scores of about 74% and 76%, respectively. Simulations also showed that the ANN-model could reach a training accuracy score of about 98% with 60 epochs. On the other hand, the RF-based ERC model was able to predict emotions with the best accuracy and precision scores of about 78% and 75%, respectively.
comment: 10 pages, 3 figures, 4 tables
☆ Stubborn Lexical Bias in Data and Models ACL
In NLP, recent work has seen increased focus on spurious correlations between various features and labels in training data, and how these influence model behavior. However, the presence and effect of such correlations are typically examined feature by feature. We investigate the cumulative impact on a model of many such intersecting features. Using a new statistical method, we examine whether such spurious patterns in data appear in models trained on the data. We select two tasks -- natural language inference and duplicate-question detection -- for which any unigram feature on its own should ideally be uninformative, which gives us a large pool of automatically extracted features with which to experiment. The large size of this pool allows us to investigate the intersection of features spuriously associated with (potentially different) labels. We then apply an optimization approach to *reweight* the training data, reducing thousands of spurious correlations, and examine how doing so affects models trained on the reweighted data. Surprisingly, though this method can successfully reduce lexical biases in the training data, we still find strong evidence of corresponding bias in the trained models, including worsened bias for slightly more complex features (bigrams). We close with discussion about the implications of our results on what it means to "debias" training data, and how issues of data quality can affect model bias.
comment: ACL Findings 2023
☆ FlairNLP at SemEval-2023 Task 6b: Extraction of Legal Named Entities from Legal Texts using Contextual String Embeddings
Indian court legal texts and processes are essential towards the integrity of the judicial system and towards maintaining the social and political order of the nation. Due to the increase in number of pending court cases, there is an urgent need to develop tools to automate many of the legal processes with the knowledge of artificial intelligence. In this paper, we employ knowledge extraction techniques, specially the named entity extraction of legal entities within court case judgements. We evaluate several state of the art architectures in the realm of sequence labeling using models trained on a curated dataset of legal texts. We observe that a Bi-LSTM model trained on Flair Embeddings achieves the best results, and we also publish the BIO formatted dataset as part of this paper.
comment: 5 pages, 4 figures
☆ TART: Improved Few-shot Text Classification Using Task-Adaptive Reference Transformation ACL 2023
Meta-learning has emerged as a trending technique to tackle few-shot text classification and achieve state-of-the-art performance. However, the performance of existing approaches heavily depends on the inter-class variance of the support set. As a result, it can perform well on tasks when the semantics of sampled classes are distinct while failing to differentiate classes with similar semantics. In this paper, we propose a novel Task-Adaptive Reference Transformation (TART) network, aiming to enhance the generalization by transforming the class prototypes to per-class fixed reference points in task-adaptive metric spaces. To further maximize divergence between transformed prototypes in task-adaptive metric spaces, TART introduces a discriminative reference regularization among transformed prototypes. Extensive experiments are conducted on four benchmark datasets and our method demonstrates clear superiority over the state-of-the-art models in all the datasets. In particular, our model surpasses the state-of-the-art method by 7.4% and 5.4% in 1-shot and 5-shot classification on the 20 Newsgroups dataset, respectively.
comment: 11 pages, 5 figures. Accepted by ACL 2023
☆ Acoustic Word Embeddings for Untranscribed Target Languages with Continued Pretraining and Learned Pooling
Acoustic word embeddings are typically created by training a pooling function using pairs of word-like units. For unsupervised systems, these are mined using k-nearest neighbor (KNN) search, which is slow. Recently, mean-pooled representations from a pre-trained self-supervised English model were suggested as a promising alternative, but their performance on target languages was not fully competitive. Here, we explore improvements to both approaches: we use continued pre-training to adapt the self-supervised model to the target language, and we use a multilingual phone recognizer (MPR) to mine phone n-gram pairs for training the pooling function. Evaluating on four languages, we show that both methods outperform a recent approach on word discrimination. Moreover, the MPR method is orders of magnitude faster than KNN, and is highly data efficient. We also show a small improvement from performing learned pooling on top of the continued pre-trained representations.
comment: Accepted to Interspeech 2023
☆ Extending an Event-type Ontology: Adding Verbs and Classes Using Fine-tuned LLMs Suggestions ACL 2023
In this project, we have investigated the use of advanced machine learning methods, specifically fine-tuned large language models, for pre-annotating data for a lexical extension task, namely adding descriptive words (verbs) to an existing (but incomplete, as of yet) ontology of event types. Several research questions have been focused on, from the investigation of a possible heuristics to provide at least hints to annotators which verbs to include and which are outside the current version of the ontology, to the possible use of the automatic scores to help the annotators to be more efficient in finding a threshold for identifying verbs that cannot be assigned to any existing class and therefore they are to be used as seeds for a new class. We have also carefully examined the correlation of the automatic scores with the human annotation. While the correlation turned out to be strong, its influence on the annotation proper is modest due to its near linearity, even though the mere fact of such pre-annotation leads to relatively short annotation times.
comment: Accepted to LAW-XVII @ ACL 2023
☆ Table and Image Generation for Investigating Knowledge of Entities in Pre-trained Vision and Language Models ACL 2023
In this paper, we propose a table and image generation task to verify how the knowledge about entities acquired from natural language is retained in Vision & Language (V & L) models. This task consists of two parts: the first is to generate a table containing knowledge about an entity and its related image, and the second is to generate an image from an entity with a caption and a table containing related knowledge of the entity. In both tasks, the model must know the entities used to perform the generation properly. We created the Wikipedia Table and Image Generation (WikiTIG) dataset from about 200,000 infoboxes in English Wikipedia articles to perform the proposed tasks. We evaluated the performance on the tasks with respect to the above research question using the V & L model OFA, which has achieved state-of-the-art results in multiple tasks. Experimental results show that OFA forgets part of its entity knowledge by pre-training as a complement to improve the performance of image related tasks.
comment: Accepted at ACL 2023
☆ Adapting Pretrained ASR Models to Low-resource Clinical Speech using Epistemic Uncertainty-based Data Selection
While there has been significant progress in ASR, African-accented clinical ASR has been understudied due to a lack of training datasets. Building robust ASR systems in this domain requires large amounts of annotated or labeled data, for a wide variety of linguistically and morphologically rich accents, which are expensive to create. Our study aims to address this problem by reducing annotation expenses through informative uncertainty-based data selection. We show that incorporating epistemic uncertainty into our adaptation rounds outperforms several baseline results, established using state-of-the-art (SOTA) ASR models, while reducing the required amount of labeled data, and hence reducing annotation costs. Our approach also improves out-of-distribution generalization for very low-resource accents, demonstrating the viability of our approach for building generalizable ASR models in the context of accented African clinical ASR, where training datasets are predominantly scarce.
☆ Benchmarking Robustness of Adaptation Methods on Pre-trained Vision-Language Models
Various adaptation methods, such as LoRA, prompts, and adapters, have been proposed to enhance the performance of pre-trained vision-language models in specific domains. The robustness of these adaptation methods against distribution shifts have not been studied. In this study, we assess the robustness of 11 widely-used adaptation methods across 4 vision-language datasets under multimodal corruptions. Concretely, we introduce 7 benchmark datasets, including 96 visual and 87 textual corruptions, to investigate the robustness of different adaptation methods, the impact of available adaptation examples, and the influence of trainable parameter size during adaptation. Our analysis reveals that: 1) Adaptation methods are more sensitive to text corruptions than visual corruptions. 2) Full fine-tuning does not consistently provide the highest robustness; instead, adapters can achieve better robustness with comparable clean performance. 3) Contrary to expectations, our findings indicate that increasing the number of adaptation data and parameters does not guarantee enhanced robustness; instead it results in even lower robustness. We hope this study could benefit future research in the development of robust multimodal adaptation methods. The benchmark, code, and dataset used in this study can be accessed at \url{https://adarobustness.github.io}.
☆ Incorporating Deep Syntactic and Semantic Knowledge for Chinese Sequence Labeling with GCN
Recently, it is quite common to integrate Chinese sequence labeling results to enhance syntactic and semantic parsing. However, little attention has been paid to the utility of hierarchy and structure information encoded in syntactic and semantic features for Chinese sequence labeling tasks. In this paper, we propose a novel framework to encode syntactic structure features and semantic information for Chinese sequence labeling tasks with graph convolutional networks (GCN). Experiments on five benchmark datasets, including Chinese word segmentation and part-of-speech tagging, demonstrate that our model can effectively improve the performance of Chinese labeling tasks.
comment: 10 pages,3 Figures, 6 Tables
☆ A Conditional Generative Chatbot using Transformer Model
A Chatbot serves as a communication tool between a human user and a machine to achieve an appropriate answer based on the human input. In more recent approaches, a combination of Natural Language Processing and sequential models are used to build a generative Chatbot. The main challenge of these models is their sequential nature, which leads to less accurate results. To tackle this challenge, in this paper, a novel end-to-end architecture is proposed using conditional Wasserstein Generative Adversarial Networks and a transformer model for answer generation in Chatbots. While the generator of the proposed model consists of a full transformer model to generate an answer, the discriminator includes only the encoder part of a transformer model followed by a classifier. To the best of our knowledge, this is the first time that a generative Chatbot is proposed using the embedded transformer in both generator and discriminator models. Relying on the parallel computing of the transformer model, the results of the proposed model on the Cornell Movie-Dialog corpus and the Chit-Chat datasets confirm the superiority of the proposed model compared to state-of-the-art alternatives using different evaluation metrics.
♻ ☆ Z-ICL: Zero-Shot In-Context Learning with Pseudo-Demonstrations
Although large language models can be prompted for both zero- and few-shot learning, performance drops significantly when no demonstrations are available. In this paper, we introduce Z-ICL, a new zero-shot method that closes the gap by constructing pseudo-demonstrations for a given test input using a raw text corpus. Concretely, pseudo-demonstrations are constructed by (1) finding the nearest neighbors to the test input from the corpus and pairing them with random task labels, and (2) applying a set of techniques to reduce the amount of direct copying the model does from the resulting demonstrations. Evaluation on nine classification datasets shows that Z-ICL outperforms previous zero-shot methods by a significant margin, and is on par with in-context learning with labeled training data in the few-shot setting. Overall, Z-ICL provides a significantly higher estimate of the zero-shot performance levels of a model, and supports future efforts to develop better pseudo-demonstrations that further improve zero-shot results.
comment: 11 pages; 9 figures
♻ ☆ MMER: Multimodal Multi-task Learning for Speech Emotion Recognition
In this paper, we propose MMER, a novel Multimodal Multi-task learning approach for Speech Emotion Recognition. MMER leverages a novel multimodal network based on early-fusion and cross-modal self-attention between text and acoustic modalities and solves three novel auxiliary tasks for learning emotion recognition from spoken utterances. In practice, MMER outperforms all our baselines and achieves state-of-the-art performance on the IEMOCAP benchmark. Additionally, we conduct extensive ablation studies and results analysis to prove the effectiveness of our proposed approach.
comment: InterSpeech 2023 Main Conference
♻ ☆ UCEpic: Unifying Aspect Planning and Lexical Constraints for Generating Explanations in Recommendation KDD 2023
Personalized natural language generation for explainable recommendations plays a key role in justifying why a recommendation might match a user's interests. Existing models usually control the generation process by aspect planning. While promising, these aspect-planning methods struggle to generate specific information correctly, which prevents generated explanations from being convincing. In this paper, we claim that introducing lexical constraints can alleviate the above issues. We propose a model, UCEpic, that generates high-quality personalized explanations for recommendation results by unifying aspect planning and lexical constraints in an insertion-based generation manner. Methodologically, to ensure text generation quality and robustness to various lexical constraints, we pre-train a non-personalized text generator via our proposed robust insertion process. Then, to obtain personalized explanations under this framework of insertion-based generation, we design a method of incorporating aspect planning and personalized references into the insertion process. Hence, UCEpic unifies aspect planning and lexical constraints into one framework and generates explanations for recommendations under different settings. Compared to previous recommendation explanation generators controlled by only aspects, UCEpic incorporates specific information from keyphrases and then largely improves the diversity and informativeness of generated explanations for recommendations on datasets such as RateBeer and Yelp.
comment: Accepted to KDD 2023
♻ ☆ WangLab at MEDIQA-Chat 2023: Clinical Note Generation from Doctor-Patient Conversations using Large Language Models ACL 2023
This paper describes our submission to the MEDIQA-Chat 2023 shared task for automatic clinical note generation from doctor-patient conversations. We report results for two approaches: the first fine-tunes a pre-trained language model (PLM) on the shared task data, and the second uses few-shot in-context learning (ICL) with a large language model (LLM). Both achieve high performance as measured by automatic metrics (e.g. ROUGE, BERTScore) and ranked second and first, respectively, of all submissions to the shared task. Expert human scrutiny indicates that notes generated via the ICL-based approach with GPT-4 are preferred about as often as human-written notes, making it a promising path toward automated note generation from doctor-patient conversations.
comment: Camera-ready submission to ClinicalNLP @ ACL 2023
♻ ☆ ExaRanker: Explanation-Augmented Neural Ranker
Recent work has shown that inducing a large language model (LLM) to generate explanations prior to outputting an answer is an effective strategy to improve performance on a wide range of reasoning tasks. In this work, we show that neural rankers also benefit from explanations. We use LLMs such as GPT-3.5 to augment retrieval datasets with explanations and train a sequence-to-sequence ranking model to output a relevance label and an explanation for a given query-document pair. Our model, dubbed ExaRanker, finetuned on a few thousand examples with synthetic explanations performs on par with models finetuned on 3x more examples without explanations. Furthermore, the ExaRanker model incurs no additional computational cost during ranking and allows explanations to be requested on demand.
♻ ☆ NLP Reproducibility For All: Understanding Experiences of Beginners ACL 2023
As natural language processing (NLP) has recently seen an unprecedented level of excitement, and more people are eager to enter the field, it is unclear whether current research reproducibility efforts are sufficient for this group of beginners to apply the latest developments. To understand their needs, we conducted a study with 93 students in an introductory NLP course, where students reproduced the results of recent NLP papers. Surprisingly, we find that their programming skill and comprehension of research papers have a limited impact on their effort spent completing the exercise. Instead, we find accessibility efforts by research authors to be the key to success, including complete documentation, better coding practice, and easier access to data files. Going forward, we recommend that NLP researchers pay close attention to these simple aspects of open-sourcing their work, and use insights from beginners' feedback to provide actionable ideas on how to better support them.
comment: ACL 2023 Theme Track
♻ ☆ Unified Generative & Dense Retrieval for Query Rewriting in Sponsored Search
Sponsored search is a key revenue source for search engines, where advertisers bid on keywords to target users or search queries of interest. However, finding relevant keywords for a given query is challenging due to the large and dynamic keyword space, ambiguous user/advertiser intents, and diverse possible topics and languages. In this work, we present a comprehensive comparison between two paradigms for online query rewriting: Generative (NLG) and Dense Retrieval (DR) methods. We observe that both methods offer complementary benefits that are additive. As a result, we show that around 40% of the high-quality keywords retrieved by the two approaches are unique and not retrieved by the other. To leverage the strengths of both methods, we propose CLOVER-Unity, a novel approach that unifies generative and dense retrieval methods in one single model. Through offline experiments, we show that the NLG and DR components of CLOVER-Unity consistently outperform individually trained NLG and DR models on public and internal benchmarks. Furthermore, we show that CLOVER-Unity achieves 9.8% higher good keyword density than the ensemble of two separate DR and NLG models while reducing computational costs by almost half. We conduct extensive online A/B experiments on Microsoft Bing in 140+ countries and achieve improved user engagement, with an average increase in total clicks by 0.89% and increased revenue by 1.27%. We also share our practical lessons and optimization tricks for deploying such unified models in production.
comment: 7 pages, 4 figures and 6 tables
Information Retrieval 8
☆ Utilizing ChatGPT to Enhance Clinical Trial Enrollment
Clinical trials are a critical component of evaluating the effectiveness of new medical interventions and driving advancements in medical research. Therefore, timely enrollment of patients is crucial to prevent delays or premature termination of trials. In this context, Electronic Health Records (EHRs) have emerged as a valuable tool for identifying and enrolling eligible participants. In this study, we propose an automated approach that leverages ChatGPT, a large language model, to extract patient-related information from unstructured clinical notes and generate search queries for retrieving potentially eligible clinical trials. Our empirical evaluation, conducted on two benchmark retrieval collections, shows improved retrieval performance compared to existing approaches when several general-purposed and task-specific prompts are used. Notably, ChatGPT-generated queries also outperform human-generated queries in terms of retrieval performance. These findings highlight the potential use of ChatGPT to enhance clinical trial enrollment while ensuring the quality of medical service and minimizing direct risks to patients.
comment: Under Review
♻ ☆ ExaRanker: Explanation-Augmented Neural Ranker
Recent work has shown that inducing a large language model (LLM) to generate explanations prior to outputting an answer is an effective strategy to improve performance on a wide range of reasoning tasks. In this work, we show that neural rankers also benefit from explanations. We use LLMs such as GPT-3.5 to augment retrieval datasets with explanations and train a sequence-to-sequence ranking model to output a relevance label and an explanation for a given query-document pair. Our model, dubbed ExaRanker, finetuned on a few thousand examples with synthetic explanations performs on par with models finetuned on 3x more examples without explanations. Furthermore, the ExaRanker model incurs no additional computational cost during ranking and allows explanations to be requested on demand.
♻ ☆ The hypergeometric test performs comparably to TF-IDF on standard text analysis tasks
Term frequency-inverse document frequency, or TF-IDF for short, and its many variants form a class of term weighting functions the members of which are widely used in text analysis applications. While TF-IDF was originally proposed as a heuristic, theoretical justifications grounded in information theory, probability, and the divergence from randomness paradigm have been advanced. In this work, we present an empirical study showing that TF-IDF corresponds very nearly with the hypergeometric test of statistical significance on selected real-data document retrieval, summarization, and classification tasks. These findings suggest that a fundamental mathematical connection between TF-IDF and the negative logarithm of the hypergeometric test P-value (i.e., a hypergeometric distribution tail probability) remains to be elucidated. We advance the empirical analyses herein as a first step toward explaining the long-standing effectiveness of TF-IDF from a statistical significance testing lens. It is our aspiration that these results will open the door to the systematic evaluation of significance testing derived term weighting functions in text analysis applications.
comment: 22 pages, 1 figure, 3 tables, major revisions
♻ ☆ Graph Meets LLM: A Novel Approach to Collaborative Filtering for Robust Conversational Understanding
Conversational AI systems such as Alexa need to understand defective queries to ensure robust conversational understanding and reduce user friction. These defective queries often arise from user ambiguities, mistakes, or errors in automatic speech recognition (ASR) and natural language understanding (NLU). Personalized query rewriting is an approach that focuses on reducing defects in queries by taking into account the user's individual behavior and preferences. It typically relies on an index of past successful user interactions with the conversational AI. However, unseen interactions within the user's history present additional challenges for personalized query rewriting. This paper presents our "Collaborative Query Rewriting" approach, which specifically addresses the task of rewriting new user interactions that have not been previously observed in the user's history. This approach builds a "User Feedback Interaction Graph" (FIG) of historical user-entity interactions and leverages multi-hop graph traversal to enrich each user's index to cover future unseen defective queries. The enriched user index is called a Collaborative User Index and contains hundreds of additional entries. To counteract precision degradation from the enlarged index, we add additional transformer layers to the L1 retrieval model and incorporate graph-based and guardrail features into the L2 ranking model. Since the user index can be pre-computed, we further investigate the utilization of a Large Language Model (LLM) to enhance the FIG for user-entity link prediction in the Video/Music domains. Specifically, this paper investigates the Dolly-V2 7B model. We found that the user index augmented by the fine-tuned Dolly-V2 generation significantly enhanced the coverage of future unseen user interactions, thereby boosting QR performance on unseen queries compared with the graph traversal only approach.
♻ ☆ Unified Generative & Dense Retrieval for Query Rewriting in Sponsored Search
Sponsored search is a key revenue source for search engines, where advertisers bid on keywords to target users or search queries of interest. However, finding relevant keywords for a given query is challenging due to the large and dynamic keyword space, ambiguous user/advertiser intents, and diverse possible topics and languages. In this work, we present a comprehensive comparison between two paradigms for online query rewriting: Generative (NLG) and Dense Retrieval (DR) methods. We observe that both methods offer complementary benefits that are additive. As a result, we show that around 40% of the high-quality keywords retrieved by the two approaches are unique and not retrieved by the other. To leverage the strengths of both methods, we propose CLOVER-Unity, a novel approach that unifies generative and dense retrieval methods in one single model. Through offline experiments, we show that the NLG and DR components of CLOVER-Unity consistently outperform individually trained NLG and DR models on public and internal benchmarks. Furthermore, we show that CLOVER-Unity achieves 9.8% higher good keyword density than the ensemble of two separate DR and NLG models while reducing computational costs by almost half. We conduct extensive online A/B experiments on Microsoft Bing in 140+ countries and achieve improved user engagement, with an average increase in total clicks by 0.89% and increased revenue by 1.27%. We also share our practical lessons and optimization tricks for deploying such unified models in production.
comment: 7 pages, 4 figures and 6 tables
♻ ☆ Class Anchor Margin Loss for Content-Based Image Retrieval
The performance of neural networks in content-based image retrieval (CBIR) is highly influenced by the chosen loss (objective) function. The majority of objective functions for neural models can be divided into metric learning and statistical learning. Metric learning approaches require a pair mining strategy that often lacks efficiency, while statistical learning approaches are not generating highly compact features due to their indirect feature optimization. To this end, we propose a novel repeller-attractor loss that falls in the metric learning paradigm, yet directly optimizes for the L2 metric without the need of generating pairs. Our loss is formed of three components. One leading objective ensures that the learned features are attracted to each designated learnable class anchor. The second loss component regulates the anchors and forces them to be separable by a margin, while the third objective ensures that the anchors do not collapse to zero. Furthermore, we develop a more efficient two-stage retrieval system by harnessing the learned class anchors during the first stage of the retrieval process, eliminating the need of comparing the query with every image in the database. We establish a set of four datasets (CIFAR-100, Food-101, SVHN, and Tiny ImageNet) and evaluate the proposed objective in the context of few-shot and full-set training on the CBIR task, by using both convolutional and transformer architectures. Compared to existing objective functions, our empirical evidence shows that the proposed objective is generating superior and more consistent results.
♻ ☆ Criteria Tell You More than Ratings: Criteria Preference-Aware Light Graph Convolution for Effective Multi-Criteria Recommendation KDD
The multi-criteria (MC) recommender system, which leverages MC rating information in a wide range of e-commerce areas, is ubiquitous nowadays. Surprisingly, although graph neural networks (GNNs) have been widely applied to develop various recommender systems due to GNN's high expressive capability in learning graph representations, it has been still unexplored how to design MC recommender systems with GNNs. In light of this, we make the first attempt towards designing a GNN-aided MC recommender system. Specifically, rather than straightforwardly adopting existing GNN-based recommendation methods, we devise a novel criteria preference-aware light graph convolution CPA-LGC method, which is capable of precisely capturing the criteria preference of users as well as the collaborative signal in complex high-order connectivities. To this end, we first construct an MC expansion graph that transforms user--item MC ratings into an expanded bipartite graph to potentially learn from the collaborative signal in MC ratings. Next, to strengthen the capability of criteria preference awareness, CPA-LGC incorporates newly characterized embeddings, including user-specific criteria-preference embeddings and item-specific criterion embeddings, into our graph convolution model. Through comprehensive evaluations using four real-world datasets, we demonstrate (a) the superiority over benchmark MC recommendation methods and benchmark recommendation methods using GNNs with tremendous gains, (b) the effectiveness of core components in CPA-LGC, and (c) the computational efficiency.
comment: 12 pages, 10 figures, 5 tables; 29th ACM SIGKDD Conference on Knowledge Discovery & Data (KDD 2023) (to appear) (Please cite our conference version.)
♻ ☆ Discovering Personalized Semantics for Soft Attributes in Recommender Systems using Concept Activation Vectors
Interactive recommender systems have emerged as a promising paradigm to overcome the limitations of the primitive user feedback used by traditional recommender systems (e.g., clicks, item consumption, ratings). They allow users to express intent, preferences, constraints, and contexts in a richer fashion, often using natural language (including faceted search and dialogue). Yet more research is needed to find the most effective ways to use this feedback. One challenge is inferring a user's semantic intent from the open-ended terms or attributes often used to describe a desired item, and using it to refine recommendation results. Leveraging concept activation vectors (CAVs) [26], a recently developed approach for model interpretability in machine learning, we develop a framework to learn a representation that captures the semantics of such attributes and connects them to user preferences and behaviors in recommender systems. One novel feature of our approach is its ability to distinguish objective and subjective attributes (both subjectivity of degree and of sense), and associate different senses of subjective attributes with different users. We demonstrate on both synthetic and real-world data sets that our CAV representation not only accurately interprets users' subjective semantics, but can also be used to improve recommendations through interactive item critiquing.
Computation and Language 72
☆ Multilingual Conceptual Coverage in Text-to-Image Models ACL 2023
We propose "Conceptual Coverage Across Languages" (CoCo-CroLa), a technique for benchmarking the degree to which any generative text-to-image system provides multilingual parity to its training language in terms of tangible nouns. For each model we can assess "conceptual coverage" of a given target language relative to a source language by comparing the population of images generated for a series of tangible nouns in the source language to the population of images generated for each noun under translation in the target language. This technique allows us to estimate how well-suited a model is to a target language as well as identify model-specific weaknesses, spurious correlations, and biases without a-priori assumptions. We demonstrate how it can be used to benchmark T2I models in terms of multilinguality, and how despite its simplicity it is a good proxy for impressive generalization.
comment: ACL 2023 main conference; 16 pages, 13 figures
☆ DocFormerv2: Local Features for Document Understanding
We propose DocFormerv2, a multi-modal transformer for Visual Document Understanding (VDU). The VDU domain entails understanding documents (beyond mere OCR predictions) e.g., extracting information from a form, VQA for documents and other tasks. VDU is challenging as it needs a model to make sense of multiple modalities (visual, language and spatial) to make a prediction. Our approach, termed DocFormerv2 is an encoder-decoder transformer which takes as input - vision, language and spatial features. DocFormerv2 is pre-trained with unsupervised tasks employed asymmetrically i.e., two novel document tasks on encoder and one on the auto-regressive decoder. The unsupervised tasks have been carefully designed to ensure that the pre-training encourages local-feature alignment between multiple modalities. DocFormerv2 when evaluated on nine datasets shows state-of-the-art performance over strong baselines e.g. TabFact (4.3%), InfoVQA (1.4%), FUNSD (1%). Furthermore, to show generalization capabilities, on three VQA tasks involving scene-text, Doc- Formerv2 outperforms previous comparably-sized models and even does better than much larger models (such as GIT2, PaLi and Flamingo) on some tasks. Extensive ablations show that due to its pre-training, DocFormerv2 understands multiple modalities better than prior-art in VDU.
☆ Improving Generalization in Task-oriented Dialogues with Workflows and Action Plans
Task-oriented dialogue is difficult in part because it involves understanding user intent, collecting information from the user, executing API calls, and generating helpful and fluent responses. However, for complex tasks one must also correctly do all of these things over multiple steps, and in a specific order. While large pre-trained language models can be fine-tuned end-to-end to create multi-step task-oriented dialogue agents that generate fluent text, our experiments confirm that this approach alone cannot reliably perform new multi-step tasks that are unseen during training. To address these limitations, we augment the dialogue contexts given to \textmd{text2text} transformers with known \textit{valid workflow names} and \textit{action plans}. Action plans consist of sequences of actions required to accomplish a task, and are encoded as simple sequences of keywords (e.g. verify-identity, pull-up-account, reset-password, etc.). We perform extensive experiments on the Action-Based Conversations Dataset (ABCD) with T5-small, base and large models, and show that such models: a) are able to more readily generalize to unseen workflows by following the provided plan, and b) are able to generalize to executing unseen actions if they are provided in the plan. In contrast, models are unable to fully accomplish new multi-step tasks when they are not provided action plan information, even when given new valid workflow names.
☆ Distilling Efficient Language-Specific Models for Cross-Lingual Transfer ACL 2023
Massively multilingual Transformers (MMTs), such as mBERT and XLM-R, are widely used for cross-lingual transfer learning. While these are pretrained to represent hundreds of languages, end users of NLP systems are often interested only in individual languages. For such purposes, the MMTs' language coverage makes them unnecessarily expensive to deploy in terms of model size, inference time, energy, and hardware cost. We thus propose to extract compressed, language-specific models from MMTs which retain the capacity of the original MMTs for cross-lingual transfer. This is achieved by distilling the MMT bilingually, i.e., using data from only the source and target language of interest. Specifically, we use a two-phase distillation approach, termed BiStil: (i) the first phase distils a general bilingual model from the MMT, while (ii) the second, task-specific phase sparsely fine-tunes the bilingual "student" model using a task-tuned variant of the original MMT as its "teacher". We evaluate this distillation technique in zero-shot cross-lingual transfer across a number of standard cross-lingual benchmarks. The key results indicate that the distilled models exhibit minimal degradation in target language performance relative to the base MMT despite being significantly smaller and faster. Furthermore, we find that they outperform multilingually distilled models such as DistilmBERT and MiniLMv2 while having a very modest training budget in comparison, even on a per-language basis. We also show that bilingual models distilled from MMTs greatly outperform bilingual models trained from scratch. Our code and models are available at https://github.com/AlanAnsell/bistil.
comment: Accepted to Findings of ACL 2023
☆ Resolving Interference When Merging Models
Transfer learning - i.e., further fine-tuning a pre-trained model on a downstream task - can confer significant advantages, including improved downstream performance, faster convergence, and better sample efficiency. These advantages have led to a proliferation of task-specific fine-tuned models, which typically can only perform a single task and do not benefit from one another. Recently, model merging techniques have emerged as a solution to combine multiple task-specific models into a single multitask model without performing additional training. However, existing merging methods often ignore the interference between parameters of different models, resulting in large performance drops when merging multiple models. In this paper, we demonstrate that prior merging techniques inadvertently lose valuable information due to two major sources of interference: (a) interference due to redundant parameter values and (b) disagreement on the sign of a given parameter's values across models. To address this, we propose our method, TrIm, Elect Sign & Merge (TIES-Merging), which introduces three novel steps when merging models: (1) resetting parameters that only changed a small amount during fine-tuning, (2) resolving sign conflicts, and (3) merging only the parameters that are in alignment with the final agreed-upon sign. We find that TIES-Merging outperforms several existing methods in diverse settings covering a range of modalities, domains, number of tasks, model sizes, architectures, and fine-tuning settings. We further analyze the impact of different types of interference on model parameters, highlight the importance of resolving sign interference. Our code is available at https://github.com/prateeky2806/ties-merging
comment: 20 Pages, 10 Figures, 12 Tables
☆ Learning Multi-step Reasoning from Arithmetic Task ACL 2023
Mathematical reasoning is regarded as a necessary ability for Language Models (LMs). Recent works demonstrate large LMs' impressive performance in solving math problems. The success is attributed to their Chain-of-Thought (CoT) reasoning abilities, i.e., the ability to decompose complex questions into step-by-step reasoning chains, but such ability seems only to emerge from models with abundant parameters. This work investigates how to incorporate relatively small LMs with the capabilities of multi-step reasoning. We propose to inject such abilities by continually pre-training LMs on a synthetic dataset MsAT, which stands for Multi-step Arithmetic Task. Our experiments on four math word problem datasets show the effectiveness of the proposed method in enhancing LMs' math reasoning abilities.
comment: ACL 2023. Code and data are available at https://github.com/TianduoWang/MsAT
☆ Fine-Grained Human Feedback Gives Better Rewards for Language Model Training
Language models (LMs) often exhibit undesirable text generation behaviors, including generating false, toxic, or irrelevant outputs. Reinforcement learning from human feedback (RLHF) - where human preference judgments on LM outputs are transformed into a learning signal - has recently shown promise in addressing these issues. However, such holistic feedback conveys limited information on long text outputs; it does not indicate which aspects of the outputs influenced user preference; e.g., which parts contain what type(s) of errors. In this paper, we use fine-grained human feedback (e.g., which sentence is false, which sub-sentence is irrelevant) as an explicit training signal. We introduce Fine-Grained RLHF, a framework that enables training and learning from reward functions that are fine-grained in two respects: (1) density, providing a reward after every segment (e.g., a sentence) is generated; and (2) incorporating multiple reward models associated with different feedback types (e.g., factual incorrectness, irrelevance, and information incompleteness). We conduct experiments on detoxification and long-form question answering to illustrate how learning with such reward functions leads to improved performance, supported by both automatic and human evaluation. Additionally, we show that LM behaviors can be customized using different combinations of fine-grained reward models. We release all data, collected human feedback, and codes at https://FineGrainedRLHF.github.io.
☆ DiffusEmp: A Diffusion Model-Based Framework with Multi-Grained Control for Empathetic Response Generation ACL 2023
Empathy is a crucial factor in open-domain conversations, which naturally shows one's caring and understanding to others. Though several methods have been proposed to generate empathetic responses, existing works often lead to monotonous empathy that refers to generic and safe expressions. In this paper, we propose to use explicit control to guide the empathy expression and design a framework DiffusEmp based on conditional diffusion language model to unify the utilization of dialogue context and attribute-oriented control signals. Specifically, communication mechanism, intent, and semantic frame are imported as multi-grained signals that control the empathy realization from coarse to fine levels. We then design a specific masking strategy to reflect the relationship between multi-grained signals and response tokens, and integrate it into the diffusion model to influence the generative process. Experimental results on a benchmark dataset EmpatheticDialogue show that our framework outperforms competitive baselines in terms of controllability, informativeness, and diversity without the loss of context-relatedness.
comment: accepted by ACL 2023 main conference (Oral)
☆ Learning from Partially Annotated Data: Example-aware Creation of Gap-filling Exercises for Language Learning
Since performing exercises (including, e.g., practice tests) forms a crucial component of learning, and creating such exercises requires non-trivial effort from the teacher. There is a great value in automatic exercise generation in digital tools in education. In this paper, we particularly focus on automatic creation of gapfilling exercises for language learning, specifically grammar exercises. Since providing any annotation in this domain requires human expert effort, we aim to avoid it entirely and explore the task of converting existing texts into new gap-filling exercises, purely based on an example exercise, without explicit instruction or detailed annotation of the intended grammar topics. We contribute (i) a novel neural network architecture specifically designed for aforementioned gap-filling exercise generation task, and (ii) a real-world benchmark dataset for French grammar. We show that our model for this French grammar gap-filling exercise generation outperforms a competitive baseline classifier by 8% in F1 percentage points, achieving an average F1 score of 82%. Our model implementation and the dataset are made publicly available to foster future research, thus offering a standardized evaluation and baseline solution of the proposed partially annotated data prediction task in grammar exercise creation.
comment: 12 pages, Accepted in the 18th Workshop on Innovative Use of NLP for Building Educational Applications
☆ EmoUS: Simulating User Emotions in Task-Oriented Dialogues SIGIR2023
Existing user simulators (USs) for task-oriented dialogue systems only model user behaviour on semantic and natural language levels without considering the user persona and emotions. Optimising dialogue systems with generic user policies, which cannot model diverse user behaviour driven by different emotional states, may result in a high drop-off rate when deployed in the real world. Thus, we present EmoUS, a user simulator that learns to simulate user emotions alongside user behaviour. EmoUS generates user emotions, semantic actions, and natural language responses based on the user goal, the dialogue history, and the user persona. By analysing what kind of system behaviour elicits what kind of user emotions, we show that EmoUS can be used as a probe to evaluate a variety of dialogue systems and in particular their effect on the user's emotional state. Developing such methods is important in the age of large language model chat-bots and rising ethical concerns.
comment: accepted by SIGIR2023
☆ Comparing a composite model versus chained models to locate a nearest visual object
Extracting information from geographic images and text is crucial for autonomous vehicles to determine in advance the best cell stations to connect to along their future path. Multiple artificial neural network models can address this challenge; however, there is no definitive guidance on the selection of an appropriate model for such use cases. Therefore, we experimented two architectures to solve such a task: a first architecture with chained models where each model in the chain addresses a sub-task of the task; and a second architecture with a single model that addresses the whole task. Our results showed that these two architectures achieved the same level performance with a root mean square error (RMSE) of 0.055 and 0.056; The findings further revealed that when the task can be decomposed into sub-tasks, the chain architecture exhibits a twelve-fold increase in training speed compared to the composite model. Nevertheless, the composite model significantly alleviates the burden of data labeling.
comment: 9 pages, 4 figures, 2 tables
☆ Evaluating Machine Translation Quality with Conformal Predictive Distributions
This paper presents a new approach for assessing uncertainty in machine translation by simultaneously evaluating translation quality and providing a reliable confidence score. Our approach utilizes conformal predictive distributions to produce prediction intervals with guaranteed coverage, meaning that for any given significance level $\epsilon$, we can expect the true quality score of a translation to fall out of the interval at a rate of $1-\epsilon$. In this paper, we demonstrate how our method outperforms a simple, but effective baseline on six different language pairs in terms of coverage and sharpness. Furthermore, we validate that our approach requires the data exchangeability assumption to hold for optimal performance.
comment: Accepted at the 12th Symposium on Conformal and Probabilistic Prediction with Applications, COPA 2023
☆ PassGPT: Password Modeling and (Guided) Generation with Large Language Models
Large language models (LLMs) successfully model natural language from vast amounts of text without the need for explicit supervision. In this paper, we investigate the efficacy of LLMs in modeling passwords. We present PassGPT, a LLM trained on password leaks for password generation. PassGPT outperforms existing methods based on generative adversarial networks (GAN) by guessing twice as many previously unseen passwords. Furthermore, we introduce the concept of guided password generation, where we leverage PassGPT sampling procedure to generate passwords matching arbitrary constraints, a feat lacking in current GAN-based strategies. Lastly, we conduct an in-depth analysis of the entropy and probability distribution that PassGPT defines over passwords and discuss their use in enhancing existing password strength estimators.
☆ BabySLM: language-acquisition-friendly benchmark of self-supervised spoken language models
Self-supervised techniques for learning speech representations have been shown to develop linguistic competence from exposure to speech without the need for human labels. In order to fully realize the potential of these approaches and further our understanding of how infants learn language, simulations must closely emulate real-life situations by training on developmentally plausible corpora and benchmarking against appropriate test sets. To this end, we propose a language-acquisition-friendly benchmark to probe spoken language models at the lexical and syntactic levels, both of which are compatible with the vocabulary typical of children's language experiences. This paper introduces the benchmark and summarizes a range of experiments showing its usefulness. In addition, we highlight two exciting challenges that need to be addressed for further progress: bridging the gap between text and speech and between clean speech and in-the-wild speech.
comment: Proceedings of Interspeech 2023
☆ Supervised Adversarial Contrastive Learning for Emotion Recognition in Conversations ACL 2023
Extracting generalized and robust representations is a major challenge in emotion recognition in conversations (ERC). To address this, we propose a supervised adversarial contrastive learning (SACL) framework for learning class-spread structured representations. The framework applies contrast-aware adversarial training to generate worst-case samples and uses a joint class-spread contrastive learning objective on both original and adversarial samples. It can effectively utilize label-level feature consistency and retain fine-grained intra-class features. To avoid the negative impact of adversarial perturbations on context-dependent data, we design a contextual adversarial training strategy to learn more diverse features from context and enhance the model's context robustness. We develop a sequence-based method SACL-LSTM under this framework, to learn label-consistent and context-robust emotional features for ERC. Experiments on three datasets demonstrate that SACL-LSTM achieves state-of-the-art performance on ERC. Extended experiments prove the effectiveness of the SACL framework.
comment: 16 pages, accepted by ACL 2023
☆ Can LLMs like GPT-4 outperform traditional AI tools in dementia diagnosis? Maybe, but not today
Recent investigations show that large language models (LLMs), specifically GPT-4, not only have remarkable capabilities in common Natural Language Processing (NLP) tasks but also exhibit human-level performance on various professional and academic benchmarks. However, whether GPT-4 can be directly used in practical applications and replace traditional artificial intelligence (AI) tools in specialized domains requires further experimental validation. In this paper, we explore the potential of LLMs such as GPT-4 to outperform traditional AI tools in dementia diagnosis. Comprehensive comparisons between GPT-4 and traditional AI tools are conducted to examine their diagnostic accuracy in a clinical setting. Experimental results on two real clinical datasets show that, although LLMs like GPT-4 demonstrate potential for future advancements in dementia diagnosis, they currently do not surpass the performance of traditional AI tools. The interpretability and faithfulness of GPT-4 are also evaluated by comparison with real doctors. We discuss the limitations of GPT-4 in its current state and propose future research directions to enhance GPT-4 in dementia diagnosis.
comment: 16 pages, 6 figures
☆ Data-Efficient French Language Modeling with CamemBERTa ACL 2023
Recent advances in NLP have significantly improved the performance of language models on a variety of tasks. While these advances are largely driven by the availability of large amounts of data and computational power, they also benefit from the development of better training methods and architectures. In this paper, we introduce CamemBERTa, a French DeBERTa model that builds upon the DeBERTaV3 architecture and training objective. We evaluate our model's performance on a variety of French downstream tasks and datasets, including question answering, part-of-speech tagging, dependency parsing, named entity recognition, and the FLUE benchmark, and compare against CamemBERT, the state-of-the-art monolingual model for French. Our results show that, given the same amount of training tokens, our model outperforms BERT-based models trained with MLM on most tasks. Furthermore, our new model reaches similar or superior performance on downstream tasks compared to CamemBERT, despite being trained on only 30% of its total number of input tokens. In addition to our experimental results, we also publicly release the weights and code implementation of CamemBERTa, making it the first publicly available DeBERTaV3 model outside of the original paper and the first openly available implementation of a DeBERTaV3 training objective. https://gitlab.inria.fr/almanach/CamemBERTa
comment: Findings of the Association for Computational Linguistics: ACL 2023, Toronto, Canda
☆ GAIA Search: Hugging Face and Pyserini Interoperability for NLP Training Data Exploration
Noticing the urgent need to provide tools for fast and user-friendly qualitative analysis of large-scale textual corpora of the modern NLP, we propose to turn to the mature and well-tested methods from the domain of Information Retrieval (IR) - a research field with a long history of tackling TB-scale document collections. We discuss how Pyserini - a widely used toolkit for reproducible IR research can be integrated with the Hugging Face ecosystem of open-source AI libraries and artifacts. We leverage the existing functionalities of both platforms while proposing novel features further facilitating their integration. Our goal is to give NLP researchers tools that will allow them to develop retrieval-based instrumentation for their data analytics needs with ease and agility. We include a Jupyter Notebook-based walk through the core interoperability features, available on GitHub at https://github.com/huggingface/gaia. We then demonstrate how the ideas we present can be operationalized to create a powerful tool for qualitative data analysis in NLP. We present GAIA Search - a search engine built following previously laid out principles, giving access to four popular large-scale text collections. GAIA serves a dual purpose of illustrating the potential of methodologies we discuss but also as a standalone qualitative analysis tool that can be leveraged by NLP researchers aiming to understand datasets prior to using them in training. GAIA is hosted live on Hugging Face Spaces - https://huggingface.co/spaces/spacerini/gaia.
☆ Guiding Text-to-Text Privatization by Syntax
Metric Differential Privacy is a generalization of differential privacy tailored to address the unique challenges of text-to-text privatization. By adding noise to the representation of words in the geometric space of embeddings, words are replaced with words located in the proximity of the noisy representation. Since embeddings are trained based on word co-occurrences, this mechanism ensures that substitutions stem from a common semantic context. Without considering the grammatical category of words, however, this mechanism cannot guarantee that substitutions play similar syntactic roles. We analyze the capability of text-to-text privatization to preserve the grammatical category of words after substitution and find that surrogate texts consist almost exclusively of nouns. Lacking the capability to produce surrogate texts that correlate with the structure of the sensitive texts, we encompass our analysis by transforming the privatization step into a candidate selection problem in which substitutions are directed to words with matching grammatical properties. We demonstrate a substantial improvement in the performance of downstream tasks by up to $4.66\%$ while retaining comparative privacy guarantees.
☆ Light Coreference Resolution for Russian with Hierarchical Discourse Features
Coreference resolution is the task of identifying and grouping mentions referring to the same real-world entity. Previous neural models have mainly focused on learning span representations and pairwise scores for coreference decisions. However, current methods do not explicitly capture the referential choice in the hierarchical discourse, an important factor in coreference resolution. In this study, we propose a new approach that incorporates rhetorical information into neural coreference resolution models. We collect rhetorical features from automated discourse parses and examine their impact. As a base model, we implement an end-to-end span-based coreference resolver using a partially fine-tuned multilingual entity-aware language model LUKE. We evaluate our method on the RuCoCo-23 Shared Task for coreference resolution in Russian. Our best model employing rhetorical distance between mentions has ranked 1st on the development set (74.6% F1) and 2nd on the test set (73.3% F1) of the Shared Task. We hope that our work will inspire further research on incorporating discourse information in neural coreference resolution models.
comment: Accepted at Dialogue-2023 conference
☆ Driving Context into Text-to-Text Privatization
\textit{Metric Differential Privacy} enables text-to-text privatization by adding calibrated noise to the vector of a word derived from an embedding space and projecting this noisy vector back to a discrete vocabulary using a nearest neighbor search. Since words are substituted without context, this mechanism is expected to fall short at finding substitutes for words with ambiguous meanings, such as \textit{'bank'}. To account for these ambiguous words, we leverage a sense embedding and incorporate a sense disambiguation step prior to noise injection. We encompass our modification to the privatization mechanism with an estimation of privacy and utility. For word sense disambiguation on the \textit{Words in Context} dataset, we demonstrate a substantial increase in classification accuracy by $6.05\%$.
☆ Unsupervised Extractive Summarization of Emotion Triggers ACL 2023
Understanding what leads to emotions during large-scale crises is important as it can provide groundings for expressed emotions and subsequently improve the understanding of ongoing disasters. Recent approaches trained supervised models to both detect emotions and explain emotion triggers (events and appraisals) via abstractive summarization. However, obtaining timely and qualitative abstractive summaries is expensive and extremely time-consuming, requiring highly-trained expert annotators. In time-sensitive, high-stake contexts, this can block necessary responses. We instead pursue unsupervised systems that extract triggers from text. First, we introduce CovidET-EXT, augmenting (Zhan et al. 2022)'s abstractive dataset (in the context of the COVID-19 crisis) with extractive triggers. Second, we develop new unsupervised learning models that can jointly detect emotions and summarize their triggers. Our best approach, entitled Emotion-Aware Pagerank, incorporates emotion information from external sources combined with a language understanding module, and outperforms strong baselines. We release our data and code at https://github.com/tsosea2/CovidET-EXT.
comment: ACL 2023 Camera-Ready
☆ Unsupervised Paraphrasing of Multiword Expressions ACL 2023
We propose an unsupervised approach to paraphrasing multiword expressions (MWEs) in context. Our model employs only monolingual corpus data and pre-trained language models (without fine-tuning), and does not make use of any external resources such as dictionaries. We evaluate our method on the SemEval 2022 idiomatic semantic text similarity task, and show that it outperforms all unsupervised systems and rivals supervised systems.
comment: 13 pages; accepted for Findings of ACL 2023
☆ Towards Robust FastSpeech 2 by Modelling Residual Multimodality INTERSPEECH 2023
State-of-the-art non-autoregressive text-to-speech (TTS) models based on FastSpeech 2 can efficiently synthesise high-fidelity and natural speech. For expressive speech datasets however, we observe characteristic audio distortions. We demonstrate that such artefacts are introduced to the vocoder reconstruction by over-smooth mel-spectrogram predictions, which are induced by the choice of mean-squared-error (MSE) loss for training the mel-spectrogram decoder. With MSE loss FastSpeech 2 is limited to learn conditional averages of the training distribution, which might not lie close to a natural sample if the distribution still appears multimodal after all conditioning signals. To alleviate this problem, we introduce TVC-GMM, a mixture model of Trivariate-Chain Gaussian distributions, to model the residual multimodality. TVC-GMM reduces spectrogram smoothness and improves perceptual audio quality in particular for expressive datasets as shown by both objective and subjective evaluation.
comment: Accepted at INTERSPEECH 2023
☆ Interpretable and Explainable Logical Policies via Neurally Guided Symbolic Abstraction
The limited priors required by neural networks make them the dominating choice to encode and learn policies using reinforcement learning (RL). However, they are also black-boxes, making it hard to understand the agent's behaviour, especially when working on the image level. Therefore, neuro-symbolic RL aims at creating policies that are interpretable in the first place. Unfortunately, interpretability is not explainability. To achieve both, we introduce Neurally gUided Differentiable loGic policiEs (NUDGE). NUDGE exploits trained neural network-based agents to guide the search of candidate-weighted logic rules, then uses differentiable logic to train the logic agents. Our experimental evaluation demonstrates that NUDGE agents can induce interpretable and explainable policies while outperforming purely neural ones and showing good flexibility to environments of different initial states and problem sizes.
comment: 9 main pages + appendix (19 in total)
☆ Knowledge Graph Reasoning over Entities and Numerical Values
A complex logic query in a knowledge graph refers to a query expressed in logic form that conveys a complex meaning, such as where did the Canadian Turing award winner graduate from? Knowledge graph reasoning-based applications, such as dialogue systems and interactive search engines, rely on the ability to answer complex logic queries as a fundamental task. In most knowledge graphs, edges are typically used to either describe the relationships between entities or their associated attribute values. An attribute value can be in categorical or numerical format, such as dates, years, sizes, etc. However, existing complex query answering (CQA) methods simply treat numerical values in the same way as they treat entities. This can lead to difficulties in answering certain queries, such as which Australian Pulitzer award winner is born before 1927, and which drug is a pain reliever and has fewer side effects than Paracetamol. In this work, inspired by the recent advances in numerical encoding and knowledge graph reasoning, we propose numerical complex query answering. In this task, we introduce new numerical variables and operations to describe queries involving numerical attribute values. To address the difference between entities and numerical values, we also propose the framework of Number Reasoning Network (NRN) for alternatively encoding entities and numerical values into separate encoding structures. During the numerical encoding process, NRN employs a parameterized density function to encode the distribution of numerical values. During the entity encoding process, NRN uses established query encoding methods for the original CQA problem. Experimental results show that NRN consistently improves various query encoding methods on three different knowledge graphs and achieves state-of-the-art results.
☆ Assessing the Importance of Frequency versus Compositionality for Subword-based Tokenization in NMT
Subword tokenization is the de facto standard for tokenization in neural language models and machine translation systems. Three advantages are frequently cited in favor of subwords: shorter encoding of frequent tokens, compositionality of subwords, and ability to deal with unknown words. As their relative importance is not entirely clear yet, we propose a tokenization approach that enables us to separate frequency (the first advantage) from compositionality. The approach uses Huffman coding to tokenize words, by order of frequency, using a fixed amount of symbols. Experiments with CS-DE, EN-FR and EN-DE NMT show that frequency alone accounts for 90%-95% of the scores reached by BPE, hence compositionality has less importance than previously thought.
comment: Accepted at EAMT 2023
☆ ChatGPT for Zero-shot Dialogue State Tracking: A Solution or an Opportunity? ACL 2023
Recent research on dialogue state tracking (DST) focuses on methods that allow few- and zero-shot transfer to new domains or schemas. However, performance gains heavily depend on aggressive data augmentation and fine-tuning of ever larger language model based architectures. In contrast, general purpose language models, trained on large amounts of diverse data, hold the promise of solving any kind of task without task-specific training. We present preliminary experimental results on the ChatGPT research preview, showing that ChatGPT achieves state-of-the-art performance in zero-shot DST. Despite our findings, we argue that properties inherent to general purpose models limit their ability to replace specialized systems. We further theorize that the in-context learning capabilities of such models will likely become powerful tools to support the development of dedicated and dynamic dialogue state trackers.
comment: 13 pages, 3 figures, accepted at ACL 2023
☆ Task-Agnostic Structured Pruning of Speech Representation Models INTERSPEECH 2023
Self-supervised pre-trained models such as Wav2vec2, Hubert, and WavLM have been shown to significantly improve many speech tasks. However, their large memory and strong computational requirements hinder their industrial applicability. Structured pruning is a hardware-friendly model compression technique but usually results in a larger loss of accuracy. In this paper, we propose a fine-grained attention head pruning method to compensate for the performance degradation. In addition, we also introduce the straight through estimator into the L0 regularization to further accelerate the pruned model. Experiments on the SUPERB benchmark show that our model can achieve comparable performance to the dense model in multiple tasks and outperforms the Wav2vec 2.0 base model on average, with 72% fewer parameters and 2 times faster inference speed.
comment: Accepted by INTERSPEECH 2023
☆ Leveraging Auxiliary Domain Parallel Data in Intermediate Task Fine-tuning for Low-resource Translation ICLR 2023
NMT systems trained on Pre-trained Multilingual Sequence-Sequence (PMSS) models flounder when sufficient amounts of parallel data is not available for fine-tuning. This specifically holds for languages missing/under-represented in these models. The problem gets aggravated when the data comes from different domains. In this paper, we show that intermediate-task fine-tuning (ITFT) of PMSS models is extremely beneficial for domain-specific NMT, especially when target domain data is limited/unavailable and the considered languages are missing or under-represented in the PMSS model. We quantify the domain-specific results variations using a domain-divergence test, and show that ITFT can mitigate the impact of domain divergence to some extent.
comment: Accepted for poster presentation at the Practical Machine Learning for Developing Countries (PML4DC) workshop, ICLR 2023
☆ An Empirical Study on Challenging Math Problem Solving with GPT-4
Employing Large Language Models (LLMs) to address mathematical problems is an intriguing research endeavor, considering the abundance of math problems expressed in natural language across numerous science and engineering fields. While several prior works have investigated solving elementary mathematics using LLMs, this work explores the frontier of using GPT-4 for solving more complex and challenging math problems. We evaluate various ways of using GPT-4. Some of them are adapted from existing work, and one is \MathChat, a conversational problem-solving framework newly proposed in this work. We perform the evaluation on difficult high school competition problems from the MATH dataset, which shows the advantage of the proposed conversational approach.
☆ Speech Translation with Foundation Models and Optimal Transport: UPC at IWSLT23
This paper describes the submission of the UPC Machine Translation group to the IWSLT 2023 Offline Speech Translation task. Our Speech Translation systems utilize foundation models for speech (wav2vec 2.0) and text (mBART50). We incorporate a Siamese pretraining step of the speech and text encoders with CTC and Optimal Transport, to adapt the speech representations to the space of the text model, thus maximizing transfer learning from MT. After this pretraining, we fine-tune our system end-to-end on ST, with Cross Entropy and Knowledge Distillation. Apart from the available ST corpora, we create synthetic data with SegAugment to better adapt our models to the custom segmentations of the IWSLT test sets. Our best single model obtains 31.2 BLEU points on MuST-C tst-COMMON, 29.8 points on IWLST.tst2020 and 33.4 points on the newly released IWSLT.ACLdev2023.
comment: IWSLT 2023
☆ LyricSIM: A novel Dataset and Benchmark for Similarity Detection in Spanish Song LyricS
In this paper, we present a new dataset and benchmark tailored to the task of semantic similarity in song lyrics. Our dataset, originally consisting of 2775 pairs of Spanish songs, was annotated in a collective annotation experiment by 63 native annotators. After collecting and refining the data to ensure a high degree of consensus and data integrity, we obtained 676 high-quality annotated pairs that were used to evaluate the performance of various state-of-the-art monolingual and multilingual language models. Consequently, we established baseline results that we hope will be useful to the community in all future academic and industrial applications conducted in this context.
comment: Accepted to Congreso Internacional de la Sociedad Espa\~nola para el Procesamiento del Lenguaje Natural 2023 (SEPLN2023)
☆ Text Style Transfer Back-Translation
Back Translation (BT) is widely used in the field of machine translation, as it has been proved effective for enhancing translation quality. However, BT mainly improves the translation of inputs that share a similar style (to be more specific, translation-like inputs), since the source side of BT data is machine-translated. For natural inputs, BT brings only slight improvements and sometimes even adverse effects. To address this issue, we propose Text Style Transfer Back Translation (TST BT), which uses a style transfer model to modify the source side of BT data. By making the style of source-side text more natural, we aim to improve the translation of natural inputs. Our experiments on various language pairs, including both high-resource and low-resource ones, demonstrate that TST BT significantly improves translation performance against popular BT benchmarks. In addition, TST BT is proved to be effective in domain adaptation so this strategy can be regarded as a general data augmentation method. Our training code and text style transfer model are open-sourced.
comment: acl2023, 14 pages, 4 figures, 19 tables
☆ Syntax-aware Hybrid prompt model for Few-shot multi-modal sentiment analysis
Multimodal Sentiment Analysis (MSA) has been a popular topic in natural language processing nowadays, at both sentence and aspect level. However, the existing approaches almost require large-size labeled datasets, which bring about large consumption of time and resources. Therefore, it is practical to explore the method for few-shot sentiment analysis in cross-modalities. Previous works generally execute on textual modality, using the prompt-based methods, mainly two types: hand-crafted prompts and learnable prompts. The existing approach in few-shot multi-modality sentiment analysis task has utilized both methods, separately. We further design a hybrid pattern that can combine one or more fixed hand-crafted prompts and learnable prompts and utilize the attention mechanisms to optimize the prompt encoder. The experiments on both sentence-level and aspect-level datasets prove that we get a significant outperformance.
☆ MetaVL: Transferring In-Context Learning Ability From Language Models to Vision-Language Models
Large-scale language models have shown the ability to adapt to a new task via conditioning on a few demonstrations (i.e., in-context learning). However, in the vision-language domain, most large-scale pre-trained vision-language (VL) models do not possess the ability to conduct in-context learning. How can we enable in-context learning for VL models? In this paper, we study an interesting hypothesis: can we transfer the in-context learning ability from the language domain to VL domain? Specifically, we first meta-trains a language model to perform in-context learning on NLP tasks (as in MetaICL); then we transfer this model to perform VL tasks by attaching a visual encoder. Our experiments suggest that indeed in-context learning ability can be transferred cross modalities: our model considerably improves the in-context learning capability on VL tasks and can even compensate for the size of the model significantly. On VQA, OK-VQA, and GQA, our method could outperform the baseline model while having 20 times fewer parameters.
☆ DistilXLSR: A Light Weight Cross-Lingual Speech Representation Model INTERSPEECH 2023
Multilingual self-supervised speech representation models have greatly enhanced the speech recognition performance for low-resource languages, and the compression of these huge models has also become a crucial prerequisite for their industrial application. In this paper, we propose DistilXLSR, a distilled cross-lingual speech representation model. By randomly shuffling the phonemes of existing speech, we reduce the linguistic information and distill cross-lingual models using only English data. We also design a layer-jumping initialization method to fully leverage the teacher's pre-trained weights. Experiments on 2 kinds of teacher models and 15 low-resource languages show that our method can reduce the parameters by 50% while maintaining cross-lingual representation ability. Our method is proven to be generalizable to various languages/teacher models and has the potential to improve the cross-lingual performance of the English pre-trained models.
comment: Accepted by INTERSPEECH 2023
☆ Improved Training for End-to-End Streaming Automatic Speech Recognition Model with Punctuation INTERSPEECH 2023
Punctuated text prediction is crucial for automatic speech recognition as it enhances readability and impacts downstream natural language processing tasks. In streaming scenarios, the ability to predict punctuation in real-time is particularly desirable but presents a difficult technical challenge. In this work, we propose a method for predicting punctuated text from input speech using a chunk-based Transformer encoder trained with Connectionist Temporal Classification (CTC) loss. The acoustic model trained with long sequences by concatenating the input and target sequences can learn punctuation marks attached to the end of sentences more effectively. Additionally, by combining CTC losses on the chunks and utterances, we achieved both the improved F1 score of punctuation prediction and Word Error Rate (WER).
comment: Accepted at INTERSPEECH 2023
☆ KL-Divergence Guided Temperature Sampling
Temperature sampling is a conventional approach to diversify large language model predictions. As temperature increases, the prediction becomes diverse but also vulnerable to hallucinations -- generating tokens that are sensible but not factual. One common approach to mitigate hallucinations is to provide source/grounding documents and the model is trained to produce predictions that bind to and are attributable to the provided source. It appears that there is a trade-off between diversity and attribution. To mitigate any such trade-off, we propose to relax the constraint of having a fixed temperature over decoding steps, and a mechanism to guide the dynamic temperature according to its relevance to the source through KL-divergence. Our experiments justifies the trade-off, and shows that our sampling algorithm outperforms the conventional top-k and top-p algorithms in conversational question-answering and summarization tasks.
☆ VoteTRANS: Detecting Adversarial Text without Training by Voting on Hard Labels of Transformations ACL 2023
Adversarial attacks reveal serious flaws in deep learning models. More dangerously, these attacks preserve the original meaning and escape human recognition. Existing methods for detecting these attacks need to be trained using original/adversarial data. In this paper, we propose detection without training by voting on hard labels from predictions of transformations, namely, VoteTRANS. Specifically, VoteTRANS detects adversarial text by comparing the hard labels of input text and its transformation. The evaluation demonstrates that VoteTRANS effectively detects adversarial text across various state-of-the-art attacks, models, and datasets.
comment: Findings of ACL 2023 (long paper)
☆ Automatic Translation of Hate Speech to Non-hate Speech in Social Media Texts
In this paper, we investigate the issue of hate speech by presenting a novel task of translating hate speech into non-hate speech text while preserving its meaning. As a case study, we use Spanish texts. We provide a dataset and several baselines as a starting point for further research in the task. We evaluated our baseline results using multiple metrics, including BLEU scores. The aim of this study is to contribute to the development of more effective methods for reducing the spread of hate speech in online communities.
☆ How Ready are Pre-trained Abstractive Models and LLMs for Legal Case Judgement Summarization?
Automatic summarization of legal case judgements has traditionally been attempted by using extractive summarization methods. However, in recent years, abstractive summarization models are gaining popularity since they can generate more natural and coherent summaries. Legal domain-specific pre-trained abstractive summarization models are now available. Moreover, general-domain pre-trained Large Language Models (LLMs), such as ChatGPT, are known to generate high-quality text and have the capacity for text summarization. Hence it is natural to ask if these models are ready for off-the-shelf application to automatically generate abstractive summaries for case judgements. To explore this question, we apply several state-of-the-art domain-specific abstractive summarization models and general-domain LLMs on Indian court case judgements, and check the quality of the generated summaries. In addition to standard metrics for summary quality, we check for inconsistencies and hallucinations in the summaries. We see that abstractive summarization models generally achieve slightly higher scores than extractive models in terms of standard summary evaluation metrics such as ROUGE and BLEU. However, we often find inconsistent or hallucinated information in the generated abstractive summaries. Overall, our investigation indicates that the pre-trained abstractive summarization models and LLMs are not yet ready for fully automatic deployment for case judgement summarization; rather a human-in-the-loop approach including manual checks for inconsistencies is more suitable at present.
comment: Accepted at the 3rd Workshop on Artificial Intelligence and Intelligent Assistance for Legal Professionals in the Digital Workplace (LegalAIIA 2023), in conjunction with the ICAIL 2023 conference
☆ THiFLY Research at SemEval-2023 Task 7: A Multi-granularity System for CTR-based Textual Entailment and Evidence Retrieval SemEval2023
The NLI4CT task aims to entail hypotheses based on Clinical Trial Reports (CTRs) and retrieve the corresponding evidence supporting the justification. This task poses a significant challenge, as verifying hypotheses in the NLI4CT task requires the integration of multiple pieces of evidence from one or two CTR(s) and the application of diverse levels of reasoning, including textual and numerical. To address these problems, we present a multi-granularity system for CTR-based textual entailment and evidence retrieval in this paper. Specifically, we construct a Multi-granularity Inference Network (MGNet) that exploits sentence-level and token-level encoding to handle both textual entailment and evidence retrieval tasks. Moreover, we enhance the numerical inference capability of the system by leveraging a T5-based model, SciFive, which is pre-trained on the medical corpus. Model ensembling and a joint inference method are further utilized in the system to increase the stability and consistency of inference. The system achieves f1-scores of 0.856 and 0.853 on textual entailment and evidence retrieval tasks, resulting in the best performance on both subtasks. The experimental results corroborate the effectiveness of our proposed method. Our code is publicly available at https://github.com/THUMLP/NLI4CT.
comment: Accepted by SemEval2023
☆ Responsible Task Automation: Empowering Large Language Models as Responsible Task Automators
The recent success of Large Language Models (LLMs) signifies an impressive stride towards artificial general intelligence. They have shown a promising prospect in automatically completing tasks upon user instructions, functioning as brain-like coordinators. The associated risks will be revealed as we delegate an increasing number of tasks to machines for automated completion. A big question emerges: how can we make machines behave responsibly when helping humans automate tasks as personal copilots? In this paper, we explore this question in depth from the perspectives of feasibility, completeness and security. In specific, we present Responsible Task Automation (ResponsibleTA) as a fundamental framework to facilitate responsible collaboration between LLM-based coordinators and executors for task automation with three empowered capabilities: 1) predicting the feasibility of the commands for executors; 2) verifying the completeness of executors; 3) enhancing the security (e.g., the protection of users' privacy). We further propose and compare two paradigms for implementing the first two capabilities. One is to leverage the generic knowledge of LLMs themselves via prompt engineering while the other is to adopt domain-specific learnable models. Moreover, we introduce a local memory mechanism for achieving the third capability. We evaluate our proposed ResponsibleTA on UI task automation and hope it could bring more attentions to ensuring LLMs more responsible in diverse scenarios. The research project homepage is at https://task-automation-research.github.io/responsible_task_automation.
♻ ☆ BertNet: Harvesting Knowledge Graphs with Arbitrary Relations from Pretrained Language Models ACL 2023
It is crucial to automatically construct knowledge graphs (KGs) of diverse new relations to support knowledge discovery and broad applications. Previous KG construction methods, based on either crowdsourcing or text mining, are often limited to a small predefined set of relations due to manual cost or restrictions in text corpus. Recent research proposed to use pretrained language models (LMs) as implicit knowledge bases that accept knowledge queries with prompts. Yet, the implicit knowledge lacks many desirable properties of a full-scale symbolic KG, such as easy access, navigation, editing, and quality assurance. In this paper, we propose a new approach of harvesting massive KGs of arbitrary relations from pretrained LMs. With minimal input of a relation definition (a prompt and a few shot of example entity pairs), the approach efficiently searches in the vast entity pair space to extract diverse accurate knowledge of the desired relation. We develop an effective search-and-rescore mechanism for improved efficiency and accuracy. We deploy the approach to harvest KGs of over 400 new relations from different LMs. Extensive human and automatic evaluations show our approach manages to extract diverse accurate knowledge, including tuples of complex relations (e.g., "A is capable of but not good at B"). The resulting KGs as a symbolic interpretation of the source LMs also reveal new insights into the LMs' knowledge capacities.
comment: ACL 2023 (Findings); Code available at https://github.com/tanyuqian/knowledge-harvest-from-lms
♻ ☆ ThinkSum: Probabilistic reasoning over sets using large language models ACL 2023
Large language models (LLMs) have a substantial capacity for high-level analogical reasoning: reproducing patterns in linear text that occur in their training data (zero-shot evaluation) or in the provided context (few-shot in-context learning). However, recent studies show that even the more advanced LLMs fail in scenarios that require reasoning over multiple objects or facts and making sequences of logical deductions. We propose a two-stage probabilistic inference paradigm, ThinkSum, which reasons over sets of objects or facts in a structured manner. In the first stage (Think - retrieval of associations), a LLM is queried in parallel over a set of phrases extracted from the prompt or an auxiliary model call. In the second stage (Sum - probabilistic inference or reasoning), the results of these queries are aggregated to make the final prediction. We demonstrate the possibilities and advantages of ThinkSum on the BIG-bench suite of LLM evaluation tasks, achieving improvements over the state of the art using GPT-family models on thirteen difficult tasks, often with far smaller model variants. We also compare and contrast ThinkSum with other proposed modifications to direct prompting of LLMs, such as variants of chain-of-thought prompting. Our results suggest that because the probabilistic inference in ThinkSum is performed outside of calls to the LLM, ThinkSum is less sensitive to prompt design, yields more interpretable predictions, and can be flexibly combined with latent variable models to extract structured knowledge from LLMs. Overall, our proposed paradigm represents a promising approach for enhancing the reasoning capabilities of LLMs.
comment: ACL 2023
♻ ☆ Shades of Iteration: from Elgot to Kleene
Notions of iteration range from the arguably most general Elgot iteration to a very specific Kleene iteration. The fundamental nature of Elgot iteration has been extensively explored by Bloom and Esik in the form of iteration theories, while Kleene iteration became extremely popular as an integral part of (untyped) formalisms, such as automata theory, regular expressions and Kleene algebra. Here, we establish a formal connection between Elgot iteration and Kleene iteration in the form of Elgot monads and Kleene monads, respectively. We also introduce a novel class of while-monads, which like Kleene monads admit a relatively simple description in algebraic terms. Like Elgot monads, while-monads cover a large variety of models that meaningfully support while-loops, but may fail the Kleene algebra laws, or even fail to support a Kleen iteration operator altogether.
comment: Extended version of the accepted one for "Recent Trends in Algebraic Development Techniques - 26th IFIP WG 1.3 International Workshop, WADT 2022"
♻ ☆ AfriNames: Most ASR models "butcher" African Names
Useful conversational agents must accurately capture named entities to minimize error for downstream tasks, for example, asking a voice assistant to play a track from a certain artist, initiating navigation to a specific location, or documenting a laboratory result for a patient. However, where named entities such as ``Ukachukwu`` (Igbo), ``Lakicia`` (Swahili), or ``Ingabire`` (Rwandan) are spoken, automatic speech recognition (ASR) models' performance degrades significantly, propagating errors to downstream systems. We model this problem as a distribution shift and demonstrate that such model bias can be mitigated through multilingual pre-training, intelligent data augmentation strategies to increase the representation of African-named entities, and fine-tuning multilingual ASR models on multiple African accents. The resulting fine-tuned models show an 81.5\% relative WER improvement compared with the baseline on samples with African-named entities.
comment: Accepted at Interspeech 2023 (Main Conference)
♻ ☆ REV: Information-Theoretic Evaluation of Free-Text Rationales ACL 2023
Generating free-text rationales is a promising step towards explainable NLP, yet evaluating such rationales remains a challenge. Existing metrics have mostly focused on measuring the association between the rationale and a given label. We argue that an ideal metric should focus on the new information uniquely provided in the rationale that is otherwise not provided in the input or the label. We investigate this research problem from an information-theoretic perspective using conditional V-information (Hewitt et al., 2021). More concretely, we propose a metric called REV (Rationale Evaluation with conditional V-information), to quantify the amount of new, label-relevant information in a rationale beyond the information already available in the input or the label. Experiments across four benchmarks with reasoning tasks, including chain-of-thought, demonstrate the effectiveness of REV in evaluating rationale-label pairs, compared to existing metrics. We further demonstrate REV is consistent with human judgments on rationale evaluations and provides more sensitive measurements of new information in free-text rationales. When used alongside traditional performance metrics, REV provides deeper insights into models' reasoning and prediction processes.
comment: ACL 2023
♻ ☆ PhysNLU: A Language Resource for Evaluating Natural Language Understanding and Explanation Coherence in Physics
In order for language models to aid physics research, they must first encode representations of mathematical and natural language discourse which lead to coherent explanations, with correct ordering and relevance of statements. We present a collection of datasets developed to evaluate the performance of language models in this regard, which measure capabilities with respect to sentence ordering, position, section prediction, and discourse coherence. Analysis of the data reveals equations and sub-disciplines which are most common in physics discourse, as well as the sentence-level frequency of equations and expressions. We present baselines that demonstrate how contemporary language models are challenged by coherence related tasks in physics, even when trained on mathematical natural language objectives.
♻ ☆ UKP-SQuARE: An Interactive Tool for Teaching Question Answering ACL2023
The exponential growth of question answering (QA) has made it an indispensable topic in any Natural Language Processing (NLP) course. Additionally, the breadth of QA derived from this exponential growth makes it an ideal scenario for teaching related NLP topics such as information retrieval, explainability, and adversarial attacks among others. In this paper, we introduce UKP-SQuARE as a platform for QA education. This platform provides an interactive environment where students can run, compare, and analyze various QA models from different perspectives, such as general behavior, explainability, and robustness. Therefore, students can get a first-hand experience in different QA techniques during the class. Thanks to this, we propose a learner-centered approach for QA education in which students proactively learn theoretical concepts and acquire problem-solving skills through interactive exploration, experimentation, and practical assignments, rather than solely relying on traditional lectures. To evaluate the effectiveness of UKP-SQuARE in teaching scenarios, we adopted it in a postgraduate NLP course and surveyed the students after the course. Their positive feedback shows the platform's effectiveness in their course and invites a wider adoption.
comment: Accepted by BEA workshop, ACL2023
♻ ☆ Revisiting Non-Autoregressive Translation at Scale ACL 2023
In real-world systems, scaling has been critical for improving the translation quality in autoregressive translation (AT), which however has not been well studied for non-autoregressive translation (NAT). In this work, we bridge the gap by systematically studying the impact of scaling on NAT behaviors. Extensive experiments on six WMT benchmarks over two advanced NAT models show that scaling can alleviate the commonly-cited weaknesses of NAT models, resulting in better translation performance. To reduce the side-effect of scaling on decoding speed, we empirically investigate the impact of NAT encoder and decoder on the translation performance. Experimental results on the large-scale WMT20 En-De show that the asymmetric architecture (e.g. bigger encoder and smaller decoder) can achieve comparable performance with the scaling model, while maintaining the superiority of decoding speed with standard NAT models. To this end, we establish a new benchmark by validating scaled NAT models on the scaled dataset, which can be regarded as a strong baseline for future works. We release code and system outputs at https://github.com/DeepLearnXMU/Scaling4NAT.
comment: 13 pages, Findings of ACL 2023
♻ ☆ End-to-end spoken language understanding using joint CTC loss and self-supervised, pretrained acoustic encoders ICASSP 2023
It is challenging to extract semantic meanings directly from audio signals in spoken language understanding (SLU), due to the lack of textual information. Popular end-to-end (E2E) SLU models utilize sequence-to-sequence automatic speech recognition (ASR) models to extract textual embeddings as input to infer semantics, which, however, require computationally expensive auto-regressive decoding. In this work, we leverage self-supervised acoustic encoders fine-tuned with Connectionist Temporal Classification (CTC) to extract textual embeddings and use joint CTC and SLU losses for utterance-level SLU tasks. Experiments show that our model achieves 4% absolute improvement over the the state-of-the-art (SOTA) dialogue act classification model on the DSTC2 dataset and 1.3% absolute improvement over the SOTA SLU model on the SLURP dataset.
comment: ICASSP 2023
♻ ☆ When Federated Learning Meets Pre-trained Language Models' Parameter-Efficient Tuning Methods
With increasing privacy concerns on data, recent studies have made significant progress using federated learning (FL) on privacy-sensitive natural language processing (NLP) tasks. Much literature suggests fully fine-tuning pre-trained language models (PLMs) in the FL paradigm can mitigate the data heterogeneity problem and close the performance gap with centralized training. However, large PLMs bring the curse of prohibitive communication overhead and local model adaptation costs for the FL system. To this end, we introduce various parameter-efficient tuning (PETuning) methods into federated learning. Specifically, we provide a holistic empirical study of representative PLMs tuning methods in FL. The experimental results cover the analysis of data heterogeneity levels, data scales, and different FL scenarios. Overall communication overhead can be significantly reduced by locally tuning and globally aggregating lightweight model parameters while maintaining acceptable performance in various FL settings. To facilitate the research of PETuning in FL, we also develop a federated tuning framework FedPETuning, which allows practitioners to exploit different PETuning methods under the FL training paradigm conveniently. The source code is available at \url{https://github.com/iezhuozhuo/FedETuning/tree/deltaTuning}.
♻ ☆ Linguistic Properties of Truthful Response ACL 2023
We investigate the phenomenon of an LLM's untruthful response using a large set of 220 handcrafted linguistic features. We focus on GPT-3 models and find that the linguistic profiles of responses are similar across model sizes. That is, how varying-sized LLMs respond to given prompts stays similar on the linguistic properties level. We expand upon this finding by training support vector machines that rely only upon the stylistic components of model responses to classify the truthfulness of statements. Though the dataset size limits our current findings, we show the possibility that truthfulness detection is possible without evaluating the content itself. But at the same time, the limited scope of our experiments must be taken into account in interpreting the results.
comment: TrustNLP @ ACL 2023
♻ ☆ Exploring Better Text Image Translation with Multimodal Codebook ACL 2023
Text image translation (TIT) aims to translate the source texts embedded in the image to target translations, which has a wide range of applications and thus has important research value. However, current studies on TIT are confronted with two main bottlenecks: 1) this task lacks a publicly available TIT dataset, 2) dominant models are constructed in a cascaded manner, which tends to suffer from the error propagation of optical character recognition (OCR). In this work, we first annotate a Chinese-English TIT dataset named OCRMT30K, providing convenience for subsequent studies. Then, we propose a TIT model with a multimodal codebook, which is able to associate the image with relevant texts, providing useful supplementary information for translation. Moreover, we present a multi-stage training framework involving text machine translation, image-text alignment, and TIT tasks, which fully exploits additional bilingual texts, OCR dataset and our OCRMT30K dataset to train our model. Extensive experiments and in-depth analyses strongly demonstrate the effectiveness of our proposed model and training framework.
comment: Accepted by ACL 2023 Main Conference
♻ ☆ Neural Natural Language Processing for Long Texts: A Survey of the State-of-the-Art
The adoption of Deep Neural Networks (DNNs) has greatly benefited Natural Language Processing (NLP) during the past decade. However, the demands of long document analysis are quite different from those of shorter texts, while the ever increasing size of documents uploaded on-line renders automated understanding of long texts a critical area of research. This article has two goals: a) it overviews the relevant neural building blocks, thus serving as a short tutorial, and b) it surveys the state-of-the-art in long document NLP, mainly focusing on two central tasks: document classification and document summarization. Sentiment analysis for long texts is also covered, since it is typically treated as a particular case of document classification. Additionally, this article discusses the main challenges, issues and current solutions related to long document NLP. Finally, the relevant, publicly available, annotated datasets are presented, in order to facilitate further research.
comment: 53 pages, 2 figures, 171 citations
♻ ☆ ChatGPT: Jack of all trades, master of none
OpenAI has released the Chat Generative Pre-trained Transformer (ChatGPT) and revolutionized the approach in artificial intelligence to human-model interaction. Several publications on ChatGPT evaluation test its effectiveness on well-known natural language processing (NLP) tasks. However, the existing studies are mostly non-automated and tested on a very limited scale. In this work, we examined ChatGPT's capabilities on 25 diverse analytical NLP tasks, most of them subjective even to humans, such as sentiment analysis, emotion recognition, offensiveness, and stance detection. In contrast, the other tasks require more objective reasoning like word sense disambiguation, linguistic acceptability, and question answering. We also evaluated GPT-4 model on five selected subsets of NLP tasks. We automated ChatGPT and GPT-4 prompting process and analyzed more than 49k responses. Our comparison of its results with available State-of-the-Art (SOTA) solutions showed that the average loss in quality of the ChatGPT model was about 25% for zero-shot and few-shot evaluation. For GPT-4 model, a loss for semantic tasks is significantly lower than for ChatGPT. We showed that the more difficult the task (lower SOTA performance), the higher the ChatGPT loss. It especially refers to pragmatic NLP problems like emotion recognition. We also tested the ability to personalize ChatGPT responses for selected subjective tasks via Random Contextual Few-Shot Personalization, and we obtained significantly better user-based predictions. Additional qualitative analysis revealed a ChatGPT bias, most likely due to the rules imposed on human trainers by OpenAI. Our results provide the basis for a fundamental discussion of whether the high quality of recent predictive NLP models can indicate a tool's usefulness to society and how the learning and validation procedures for such systems should be established.
comment: preprint
♻ ☆ Motion-Based Sign Language Video Summarization using Curvature and Torsion
An interesting problem in many video-based applications is the generation of short synopses by selecting the most informative frames, a procedure which is known as video summarization. For sign language videos the benefits of using the $t$-parameterized counterpart of the curvature of the 2-D signer's wrist trajectory to identify keyframes, have been recently reported in the literature. In this paper we extend these ideas by modeling the 3-D hand motion that is extracted from each frame of the video. To this end we propose a new informative function based on the $t$-parameterized curvature and torsion of the 3-D trajectory. The method to characterize video frames as keyframes depends on whether the motion occurs in 2-D or 3-D space. Specifically, in the case of 3-D motion we look for the maxima of the harmonic mean of the curvature and torsion of the target's trajectory; in the planar motion case we seek for the maxima of the trajectory's curvature. The proposed 3-D feature is experimentally evaluated in applications of sign language videos on (1) objective measures using ground-truth keyframe annotations, (2) human-based evaluation of understanding, and (3) gloss classification and the results obtained are promising.
comment: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible
♻ ☆ Lessons on Parameter Sharing across Layers in Transformers
We propose a parameter sharing method for Transformers (Vaswani et al., 2017). The proposed approach relaxes a widely used technique, which shares parameters for one layer with all layers such as Universal Transformers (Dehghani et al., 2019), to increase the efficiency in the computational time. We propose three strategies: Sequence, Cycle, and Cycle (rev) to assign parameters to each layer. Experimental results show that the proposed strategies are efficient in the parameter size and computational time. Moreover, we indicate that the proposed strategies are also effective in the configuration where we use many training data such as the recent WMT competition.
comment: SustaiNLP 2023
♻ ☆ Optimizing Non-Autoregressive Transformers with Contrastive Learning
Non-autoregressive Transformers (NATs) reduce the inference latency of Autoregressive Transformers (ATs) by predicting words all at once rather than in sequential order. They have achieved remarkable progress in machine translation as well as many other applications. However, a long-standing challenge for NATs is the learning of multi-modality data distribution, which is the main cause of the performance gap between NATs and ATs. In this paper, we propose to ease the difficulty of modality learning via sampling from the model distribution instead of the data distribution. We derive contrastive constraints to stabilize the training process and integrate this resulting objective with the state-of-the-art NAT architecture DA-Transformer. Our model \method is examined on 3 different tasks, including machine translation, text summarization, and paraphrasing with 5 benchmarks. Results show that our approach outperforms previous non-autoregressive baselines by a significant margin and establishes new state-of-the-art results for non-autoregressive transformers on all the benchmarks.
♻ ☆ Syntax-Aware Graph-to-Graph Transformer for Semantic Role Labelling ACL 2023
Recent models have shown that incorporating syntactic knowledge into the semantic role labelling (SRL) task leads to a significant improvement. In this paper, we propose Syntax-aware Graph-to-Graph Transformer (SynG2G-Tr) model, which encodes the syntactic structure using a novel way to input graph relations as embeddings, directly into the self-attention mechanism of Transformer. This approach adds a soft bias towards attention patterns that follow the syntactic structure but also allows the model to use this information to learn alternative patterns. We evaluate our model on both span-based and dependency-based SRL datasets, and outperform previous alternative methods in both in-domain and out-of-domain settings, on CoNLL 2005 and CoNLL 2009 datasets.
comment: Accepted to Rep4NLP at ACL 2023
♻ ☆ Easy Guided Decoding in Providing Suggestions for Interactive Machine Translation ACL 2023
Machine translation technology has made great progress in recent years, but it cannot guarantee error free results. Human translators perform post editing on machine translations to correct errors in the scene of computer aided translation. In favor of expediting the post editing process, many works have investigated machine translation in interactive modes, in which machines can automatically refine the rest of translations constrained by human's edits. Translation Suggestion (TS), as an interactive mode to assist human translators, requires machines to generate alternatives for specific incorrect words or phrases selected by human translators. In this paper, we utilize the parameterized objective function of neural machine translation (NMT) and propose a novel constrained decoding algorithm, namely Prefix Suffix Guided Decoding (PSGD), to deal with the TS problem without additional training. Compared to the state of the art lexically constrained decoding method, PSGD improves translation quality by an average of $10.87$ BLEU and $8.62$ BLEU on the WeTS and the WMT 2022 Translation Suggestion datasets, respectively, and reduces decoding time overhead by an average of 63.4% tested on the WMT translation datasets. Furthermore, on both of the TS benchmark datasets, it is superior to other supervised learning systems trained with TS annotated data.
comment: Accepted by ACL 2023
♻ ☆ Are You Copying My Model? Protecting the Copyright of Large Language Models for EaaS via Backdoor Watermark ACL 2023
Large language models (LLMs) have demonstrated powerful capabilities in both text understanding and generation. Companies have begun to offer Embedding as a Service (EaaS) based on these LLMs, which can benefit various natural language processing (NLP) tasks for customers. However, previous studies have shown that EaaS is vulnerable to model extraction attacks, which can cause significant losses for the owners of LLMs, as training these models is extremely expensive. To protect the copyright of LLMs for EaaS, we propose an Embedding Watermark method called EmbMarker that implants backdoors on embeddings. Our method selects a group of moderate-frequency words from a general text corpus to form a trigger set, then selects a target embedding as the watermark, and inserts it into the embeddings of texts containing trigger words as the backdoor. The weight of insertion is proportional to the number of trigger words included in the text. This allows the watermark backdoor to be effectively transferred to EaaS-stealer's model for copyright verification while minimizing the adverse impact on the original embeddings' utility. Our extensive experiments on various datasets show that our method can effectively protect the copyright of EaaS models without compromising service quality.
comment: Accepted by ACL 2023
♻ ☆ Multi-Level Knowledge Distillation for Out-of-Distribution Detection in Text ACL 2023
Self-supervised representation learning has proved to be a valuable component for out-of-distribution (OoD) detection with only the texts of in-distribution (ID) examples. These approaches either train a language model from scratch or fine-tune a pre-trained language model using ID examples, and then take the perplexity output by the language model as OoD scores. In this paper, we analyze the complementary characteristics of both OoD detection methods and propose a multi-level knowledge distillation approach that integrates their strengths while mitigating their limitations. Specifically, we use a fine-tuned model as the teacher to teach a randomly initialized student model on the ID examples. Besides the prediction layer distillation, we present a similarity-based intermediate layer distillation method to thoroughly explore the representation space of the teacher model. In this way, the learned student can better represent the ID data manifold while gaining a stronger ability to map OoD examples outside the ID data manifold with the regularization inherited from pre-training. Besides, the student model sees only ID examples during parameter learning, further promoting more distinguishable features for OoD detection. We conduct extensive experiments over multiple benchmark datasets, i.e., CLINC150, SST, ROSTD, 20 NewsGroups, and AG News; showing that the proposed method yields new state-of-the-art performance. We also explore its application as an AIGC detector to distinguish between answers generated by ChatGPT and human experts. It is observed that our model exceeds human evaluators in the pair-expert task on the Human ChatGPT Comparison Corpus.
comment: ACL 2023. Our code is available at: https://github.com/microsoft/KC/tree/main/papers/MLKD_OOD
♻ ☆ Measuring Consistency in Text-based Financial Forecasting Models ACL 2023
Financial forecasting has been an important and active area of machine learning research, as even the most modest advantage in predictive accuracy can be parlayed into significant financial gains. Recent advances in natural language processing (NLP) bring the opportunity to leverage textual data, such as earnings reports of publicly traded companies, to predict the return rate for an asset. However, when dealing with such a sensitive task, the consistency of models -- their invariance under meaning-preserving alternations in input -- is a crucial property for building user trust. Despite this, current financial forecasting methods do not consider consistency. To address this problem, we propose FinTrust, an evaluation tool that assesses logical consistency in financial text. Using FinTrust, we show that the consistency of state-of-the-art NLP models for financial forecasting is poor. Our analysis of the performance degradation caused by meaning-preserving alternations suggests that current text-based methods are not suitable for robustly predicting market information. All resources are available at https://github.com/yingpengma/fintrust.
comment: Accepted to ACL 2023 Main Conference
♻ ☆ Multiscale Positive-Unlabeled Detection of AI-Generated Texts
Recent releases of Large Language Models (LLMs), e.g. ChatGPT, are astonishing at generating human-like texts, but they may get misused for fake scholarly texts, fake news, fake tweets, et cetera. Previous works have proposed methods to detect these multiscale AI-generated texts, including simple ML classifiers, pretrained-model-based training-agnostic methods, and finetuned language classification models. However, mainstream detectors are formulated without considering the factor of corpus length: shorter corpuses are harder to detect compared with longer ones for shortage of informative features. In this paper, a Multiscale Positive-Unlabeled (MPU) training framework is proposed to address the challenge of multiscale text detection. Firstly, we acknowledge the human-resemblance property of short machine texts, and rephrase text classification as a Positive-Unlabeled (PU) problem by marking these short machine texts as "unlabeled" during training. In this PU context, we propose the length-sensitive Multiscale PU Loss, where we use a recurrent model in abstraction to estimate positive priors of scale-variant corpuses. Additionally, we introduce a Text Multiscaling module to enrich training corpuses. Experiments show that our MPU method augments detection performance on long AI-generated text, and significantly improves short-corpus detection of language model detectors. Language Models trained with MPU could outcompete existing detectors by large margins on multiscale AI-generated texts. The codes are available at https://github.com/mindspore-lab/mindone/tree/master/examples/detect_chatgpt and https://github.com/YuchuanTian/AIGC_text_detector.
♻ ☆ Unified Detoxifying and Debiasing in Language Generation via Inference-time Adaptive Optimization ICLR 2023
Warning: this paper contains model outputs exhibiting offensiveness and biases. Recently pre-trained language models (PLMs) have prospered in various natural language generation (NLG) tasks due to their ability to generate fairly fluent text. Nevertheless, these models are observed to capture and reproduce harmful contents in training corpora, typically toxic language and social biases, raising severe moral issues. Prior works on ethical NLG tackle detoxifying and debiasing separately, which is problematic since we find debiased models still exhibit toxicity while detoxified ones even exacerbate social biases. To address such a challenge, we propose the first unified framework of detoxifying and debiasing called UDDIA, which jointly formalizes these two problems as rectifying the output space. We theoretically interpret our framework as learning a text distribution mixing weighted attributes. Besides, UDDIA conducts adaptive optimization of only a few parameters during decoding based on a parameter-efficient tuning schema without any training data. This leads to minimal generation quality loss and improved rectification performance with acceptable computational cost. Experimental results demonstrate that compared to several strong baselines, UDDIA achieves debiasing and detoxifying simultaneously and better balances efficiency and effectiveness, taking a further step towards practical ethical NLG.
comment: Accepted at ICLR 2023
♻ ☆ On the Off-Target Problem of Zero-Shot Multilingual Neural Machine Translation ACL 2023
While multilingual neural machine translation has achieved great success, it suffers from the off-target issue, where the translation is in the wrong language. This problem is more pronounced on zero-shot translation tasks. In this work, we find that failing in encoding discriminative target language signal will lead to off-target and a closer lexical distance (i.e., KL-divergence) between two languages' vocabularies is related with a higher off-target rate. We also find that solely isolating the vocab of different languages in the decoder can alleviate the problem. Motivated by the findings, we propose Language Aware Vocabulary Sharing (LAVS), a simple and effective algorithm to construct the multilingual vocabulary, that greatly alleviates the off-target problem of the translation model by increasing the KL-divergence between languages. We conduct experiments on a multilingual machine translation benchmark in 11 languages. Experiments show that the off-target rate for 90 translation tasks is reduced from 29\% to 8\%, while the overall BLEU score is improved by an average of 1.9 points without extra training cost or sacrificing the supervised directions' performance. We release the code at https://github.com/PKUnlp-icler/Off-Target-MNMT for reproduction.
comment: Findings of ACL 2023
♻ ☆ TopEx: Topic-based Explanations for Model Comparison ICLR 2023
Meaningfully comparing language models is challenging with current explanation methods. Current explanations are overwhelming for humans due to large vocabularies or incomparable across models. We present TopEx, an explanation method that enables a level playing field for comparing language models via model-agnostic topics. We demonstrate how TopEx can identify similarities and differences between DistilRoBERTa and GPT-2 on a variety of NLP tasks.
comment: Accepted to ICLR 2023, Tiny Papers Track
♻ ☆ KEYword based Sampling (KEYS) for Large Language Models
Question answering (Q/A) can be formulated as a generative task (Mitra, 2017) where the task is to generate an answer given the question and the passage (knowledge, if available). Recent advances in QA task is focused a lot on language model advancements and less on other areas such as sampling(Krishna et al., 2021), (Nakano et al., 2021). Keywords play very important role for humans in language generation. (Humans formulate keywords and use grammar to connect those keywords and work). In the research community, very little focus is on how humans generate answers to a question and how this behavior can be incorporated in a language model. In this paper, we want to explore these two areas combined, i.e., how sampling can be to used generate answers which are close to human-like behavior and factually correct. Hence, the type of decoding algorithm we think should be used for Q/A tasks should also depend on the keywords. These keywords can be obtained from the question, passage or internet results. We use knowledge distillation techniques to extract keywords and sample using these extracted keywords on top of vanilla decoding algorithms when formulating the answer to generate a human-like answer. In this paper, we show that our decoding method outperforms most commonly used decoding methods for Q/A task
♻ ☆ Fluid Transformers and Creative Analogies: Exploring Large Language Models' Capacity for Augmenting Cross-Domain Analogical Creativity
Cross-domain analogical reasoning is a core creative ability that can be challenging for humans. Recent work has shown some proofs-of concept of Large language Models' (LLMs) ability to generate cross-domain analogies. However, the reliability and potential usefulness of this capacity for augmenting human creative work has received little systematic exploration. In this paper, we systematically explore LLMs capacity to augment cross-domain analogical reasoning. Across three studies, we found: 1) LLM-generated cross-domain analogies were frequently judged as helpful in the context of a problem reformulation task (median 4 out of 5 helpfulness rating), and frequently (~80% of cases) led to observable changes in problem formulations, and 2) there was an upper bound of 25% of outputs bring rated as potentially harmful, with a majority due to potentially upsetting content, rather than biased or toxic content. These results demonstrate the potential utility -- and risks -- of LLMs for augmenting cross-domain analogical creativity.
Computer Vision and Pattern Recognition 96
☆ OCBEV: Object-Centric BEV Transformer for Multi-View 3D Object Detection
Multi-view 3D object detection is becoming popular in autonomous driving due to its high effectiveness and low cost. Most of the current state-of-the-art detectors follow the query-based bird's-eye-view (BEV) paradigm, which benefits from both BEV's strong perception power and end-to-end pipeline. Despite achieving substantial progress, existing works model objects via globally leveraging temporal and spatial information of BEV features, resulting in problems when handling the challenging complex and dynamic autonomous driving scenarios. In this paper, we proposed an Object-Centric query-BEV detector OCBEV, which can carve the temporal and spatial cues of moving targets more effectively. OCBEV comprises three designs: Object Aligned Temporal Fusion aligns the BEV feature based on ego-motion and estimated current locations of moving objects, leading to a precise instance-level feature fusion. Object Focused Multi-View Sampling samples more 3D features from an adaptive local height ranges of objects for each scene to enrich foreground information. Object Informed Query Enhancement replaces part of pre-defined decoder queries in common DETR-style decoders with positional features of objects on high-confidence locations, introducing more direct object positional priors. Extensive experimental evaluations are conducted on the challenging nuScenes dataset. Our approach achieves a state-of-the-art result, surpassing the traditional BEVFormer by 1.5 NDS points. Moreover, we have a faster convergence speed and only need half of the training iterations to get comparable performance, which further demonstrates its effectiveness.
☆ DaTaSeg: Taming a Universal Multi-Dataset Multi-Task Segmentation Model
Observing the close relationship among panoptic, semantic and instance segmentation tasks, we propose to train a universal multi-dataset multi-task segmentation model: DaTaSeg.We use a shared representation (mask proposals with class predictions) for all tasks. To tackle task discrepancy, we adopt different merge operations and post-processing for different tasks. We also leverage weak-supervision, allowing our segmentation model to benefit from cheaper bounding box annotations. To share knowledge across datasets, we use text embeddings from the same semantic embedding space as classifiers and share all network parameters among datasets. We train DaTaSeg on ADE semantic, COCO panoptic, and Objects365 detection datasets. DaTaSeg improves performance on all datasets, especially small-scale datasets, achieving 54.0 mIoU on ADE semantic and 53.5 PQ on COCO panoptic. DaTaSeg also enables weakly-supervised knowledge transfer on ADE panoptic and Objects365 instance segmentation. Experiments show DaTaSeg scales with the number of training datasets and enables open-vocabulary segmentation through direct transfer. In addition, we annotate an Objects365 instance segmentation set of 1,000 images and will release it as a public benchmark.
☆ Multilingual Conceptual Coverage in Text-to-Image Models ACL 2023
We propose "Conceptual Coverage Across Languages" (CoCo-CroLa), a technique for benchmarking the degree to which any generative text-to-image system provides multilingual parity to its training language in terms of tangible nouns. For each model we can assess "conceptual coverage" of a given target language relative to a source language by comparing the population of images generated for a series of tangible nouns in the source language to the population of images generated for each noun under translation in the target language. This technique allows us to estimate how well-suited a model is to a target language as well as identify model-specific weaknesses, spurious correlations, and biases without a-priori assumptions. We demonstrate how it can be used to benchmark T2I models in terms of multilinguality, and how despite its simplicity it is a good proxy for impressive generalization.
comment: ACL 2023 main conference; 16 pages, 13 figures
☆ DocFormerv2: Local Features for Document Understanding
We propose DocFormerv2, a multi-modal transformer for Visual Document Understanding (VDU). The VDU domain entails understanding documents (beyond mere OCR predictions) e.g., extracting information from a form, VQA for documents and other tasks. VDU is challenging as it needs a model to make sense of multiple modalities (visual, language and spatial) to make a prediction. Our approach, termed DocFormerv2 is an encoder-decoder transformer which takes as input - vision, language and spatial features. DocFormerv2 is pre-trained with unsupervised tasks employed asymmetrically i.e., two novel document tasks on encoder and one on the auto-regressive decoder. The unsupervised tasks have been carefully designed to ensure that the pre-training encourages local-feature alignment between multiple modalities. DocFormerv2 when evaluated on nine datasets shows state-of-the-art performance over strong baselines e.g. TabFact (4.3%), InfoVQA (1.4%), FUNSD (1%). Furthermore, to show generalization capabilities, on three VQA tasks involving scene-text, Doc- Formerv2 outperforms previous comparably-sized models and even does better than much larger models (such as GIT2, PaLi and Flamingo) on some tasks. Extensive ablations show that due to its pre-training, DocFormerv2 understands multiple modalities better than prior-art in VDU.
☆ Video Colorization with Pre-trained Text-to-Image Diffusion Models
Video colorization is a challenging task that involves inferring plausible and temporally consistent colors for grayscale frames. In this paper, we present ColorDiffuser, an adaptation of a pre-trained text-to-image latent diffusion model for video colorization. With the proposed adapter-based approach, we repropose the pre-trained text-to-image model to accept input grayscale video frames, with the optional text description, for video colorization. To enhance the temporal coherence and maintain the vividness of colorization across frames, we propose two novel techniques: the Color Propagation Attention and Alternated Sampling Strategy. Color Propagation Attention enables the model to refine its colorization decision based on a reference latent frame, while Alternated Sampling Strategy captures spatiotemporal dependencies by using the next and previous adjacent latent frames alternatively as reference during the generative diffusion sampling steps. This encourages bidirectional color information propagation between adjacent video frames, leading to improved color consistency across frames. We conduct extensive experiments on benchmark datasets, and the results demonstrate the effectiveness of our proposed framework. The evaluations show that ColorDiffuser achieves state-of-the-art performance in video colorization, surpassing existing methods in terms of color fidelity, temporal consistency, and visual quality.
comment: project page: https://colordiffuser.github.io/
☆ Denoising Diffusion Semantic Segmentation with Mask Prior Modeling
The evolution of semantic segmentation has long been dominated by learning more discriminative image representations for classifying each pixel. Despite the prominent advancements, the priors of segmentation masks themselves, e.g., geometric and semantic constraints, are still under-explored. In this paper, we propose to ameliorate the semantic segmentation quality of existing discriminative approaches with a mask prior modeled by a recently-developed denoising diffusion generative model. Beginning with a unified architecture that adapts diffusion models for mask prior modeling, we focus this work on a specific instantiation with discrete diffusion and identify a variety of key design choices for its successful application. Our exploratory analysis revealed several important findings, including: (1) a simple integration of diffusion models into semantic segmentation is not sufficient, and a poorly-designed diffusion process might lead to degradation in segmentation performance; (2) during the training, the object to which noise is added is more important than the type of noise; (3) during the inference, the strict diffusion denoising scheme may not be essential and can be relaxed to a simpler scheme that even works better. We evaluate the proposed prior modeling with several off-the-shelf segmentors, and our experimental results on ADE20K and Cityscapes demonstrate that our approach could achieve competitively quantitative performance and more appealing visual quality.
☆ Resolving Interference When Merging Models
Transfer learning - i.e., further fine-tuning a pre-trained model on a downstream task - can confer significant advantages, including improved downstream performance, faster convergence, and better sample efficiency. These advantages have led to a proliferation of task-specific fine-tuned models, which typically can only perform a single task and do not benefit from one another. Recently, model merging techniques have emerged as a solution to combine multiple task-specific models into a single multitask model without performing additional training. However, existing merging methods often ignore the interference between parameters of different models, resulting in large performance drops when merging multiple models. In this paper, we demonstrate that prior merging techniques inadvertently lose valuable information due to two major sources of interference: (a) interference due to redundant parameter values and (b) disagreement on the sign of a given parameter's values across models. To address this, we propose our method, TrIm, Elect Sign & Merge (TIES-Merging), which introduces three novel steps when merging models: (1) resetting parameters that only changed a small amount during fine-tuning, (2) resolving sign conflicts, and (3) merging only the parameters that are in alignment with the final agreed-upon sign. We find that TIES-Merging outperforms several existing methods in diverse settings covering a range of modalities, domains, number of tasks, model sizes, architectures, and fine-tuning settings. We further analyze the impact of different types of interference on model parameters, highlight the importance of resolving sign interference. Our code is available at https://github.com/prateeky2806/ties-merging
comment: 20 Pages, 10 Figures, 12 Tables
☆ Is Generative Modeling-based Stylization Necessary for Domain Adaptation in Regression Tasks?
Unsupervised domain adaptation (UDA) aims to bridge the gap between source and target domains in the absence of target domain labels using two main techniques: input-level alignment (such as generative modeling and stylization) and feature-level alignment (which matches the distribution of the feature maps, e.g. gradient reversal layers). Motivated from the success of generative modeling for image classification, stylization-based methods were recently proposed for regression tasks, such as pose estimation. However, use of input-level alignment via generative modeling and stylization incur additional overhead and computational complexity which limit their use in real-world DA tasks. To investigate the role of input-level alignment for DA, we ask the following question: Is generative modeling-based stylization necessary for visual domain adaptation in regression? Surprisingly, we find that input-alignment has little effect on regression tasks as compared to classification. Based on these insights, we develop a non-parametric feature-level domain alignment method -- Implicit Stylization (ImSty) -- which results in consistent improvements over SOTA regression task, without the need for computationally intensive stylization and generative modeling. Our work conducts a critical evaluation of the role of generative modeling and stylization, at a time when these are also gaining popularity for domain generalization.
☆ Unique Brain Network Identification Number for Parkinson's Individuals Using Structural MRI
We propose a novel algorithm called Unique Brain Network Identification Number (UBNIN) for encoding brain networks of individual subject. To realize this objective, we employed T1-weighted structural MRI of 180 Parkinson's disease (PD) patients from National Institute of Mental Health and Neurosciences, India. We parcellated each subject's brain volume and constructed individual adjacency matrix using correlation between grey matter (GM) volume of every pair of regions. The unique code is derived from values representing connections of every node (i), weighted by a factor of 2^-(i-1). The numerical representation UBNIN was observed to be distinct for each individual brain network, which may also be applied to other neuroimaging modalities. This model may be implemented as neural signature of a person's unique brain connectivity, thereby useful for brainprinting applications. Additionally, we segregated the above dataset into five age-cohorts: A:22-32years, B:33-42years, C:43-52years, D:53-62years and E:63-72years to study the variation in network topology over age. Sparsity was adopted as the threshold estimate to binarize each age-based correlation matrix. Connectivity metrics were obtained using Brain Connectivity toolbox-based MATLAB functions. For each age-cohort, a decreasing trend was observed in mean clustering coefficient with increasing sparsity. Significantly different clustering coefficient was noted between age-cohort B and C (sparsity: 0.63,0.66), C and E (sparsity: 0.66,0.69). Our findings suggest network connectivity patterns change with age, indicating network disruption due to the underlying neuropathology. Varying clustering coefficient for different cohorts indicate that information transfer between neighboring nodes change with age. This provides evidence on age-related brain shrinkage and network degeneration.
comment: 9 pages, 5 figures,1 algorithm, 1 main table, 1 appendix table
☆ MKOR: Momentum-Enabled Kronecker-Factor-Based Optimizer Using Rank-1 Updates
This work proposes a Momentum-Enabled Kronecker-Factor-Based Optimizer Using Rank-1 updates, called MKOR, that improves the training time and convergence properties of deep neural networks (DNNs). Second-order techniques, while enjoying higher convergence rates vs first-order counterparts, have cubic complexity with respect to either the model size and/or the training batch size. Hence they exhibit poor scalability and performance in transformer models, e.g. large language models (LLMs), because the batch sizes in these models scale by the attention mechanism sequence length, leading to large model size and batch sizes. MKOR's complexity is quadratic with respect to the model size, alleviating the computation bottlenecks in second-order methods. Because of their high computation complexity, state-of-the-art implementations of second-order methods can only afford to update the second order information infrequently, and thus do not fully exploit the promise of better convergence from these updates. By reducing the communication complexity of the second-order updates as well as achieving a linear communication complexity, MKOR increases the frequency of second order updates. We also propose a hybrid version of MKOR (called MKOR-H) that mid-training falls backs to a first order optimizer if the second order updates no longer accelerate convergence. Our experiments show that MKOR outperforms state -of-the-art first order methods, e.g. the LAMB optimizer, and best implementations of second-order methods, i.e. KAISA/KFAC, up to 2.57x and 1.85x respectively on BERT-Large-Uncased on 64 GPUs.
☆ Enhancing CLIP with CLIP: Exploring Pseudolabeling for Limited-Label Prompt Tuning
Fine-tuning vision-language models (VLMs) like CLIP to downstream tasks is often necessary to optimize their performance. However, a major obstacle is the limited availability of labeled data. We study the use of pseudolabels, i.e., heuristic labels for unlabeled data, to enhance CLIP via prompt tuning. Conventional pseudolabeling trains a model on labeled data and then generates labels for unlabeled data. VLMs' zero-shot capabilities enable a ``second generation'' of pseudolabeling approaches that do not require task-specific training on labeled data. By using zero-shot pseudolabels as a source of supervision, we observe that learning paradigms such as semi-supervised, transductive zero-shot, and unsupervised learning can all be seen as optimizing the same loss function. This unified view enables the development of versatile training strategies that are applicable across learning paradigms. We investigate them on image classification tasks where CLIP exhibits limitations, by varying prompt modalities, e.g., textual or visual prompts, and learning paradigms. We find that (1) unexplored prompt tuning strategies that iteratively refine pseudolabels consistently improve CLIP accuracy, by 19.5 points in semi-supervised learning, by 28.4 points in transductive zero-shot learning, and by 15.2 points in unsupervised learning, and (2) unlike conventional semi-supervised pseudolabeling, which exacerbates model biases toward classes with higher-quality pseudolabels, prompt tuning leads to a more equitable distribution of per-class accuracy. The code to reproduce the experiments is at github.com/BatsResearch/menghini-enhanceCLIPwithCLIP-code.
☆ Towards In-context Scene Understanding
In-context learning$\unicode{x2013}$the ability to configure a model's behavior with different prompts$\unicode{x2013}$has revolutionized the field of natural language processing, alleviating the need for task-specific models and paving the way for generalist models capable of assisting with any query. Computer vision, in contrast, has largely stayed in the former regime: specialized decoders and finetuning protocols are generally required to perform dense tasks such as semantic segmentation and depth estimation. In this work we explore a simple mechanism for in-context learning of such scene understanding tasks: nearest neighbor retrieval from a prompt of annotated features. We propose a new pretraining protocol$\unicode{x2013}$leveraging attention within and across images$\unicode{x2013}$which yields representations particularly useful in this regime. The resulting Hummingbird model, suitably prompted, performs various scene understanding tasks without modification while approaching the performance of specialists that have been finetuned for each task. Moreover, Hummingbird can be configured to perform new tasks much more efficiently than finetuned models, raising the possibility of scene understanding in the interactive assistant regime.
☆ Backchannel Detection and Agreement Estimation from Video with Transformer Networks IJCNN'23
Listeners use short interjections, so-called backchannels, to signify attention or express agreement. The automatic analysis of this behavior is of key importance for human conversation analysis and interactive conversational agents. Current state-of-the-art approaches for backchannel analysis from visual behavior make use of two types of features: features based on body pose and features based on facial behavior. At the same time, transformer neural networks have been established as an effective means to fuse input from different data sources, but they have not yet been applied to backchannel analysis. In this work, we conduct a comprehensive evaluation of multi-modal transformer architectures for automatic backchannel analysis based on pose and facial information. We address both the detection of backchannels as well as the task of estimating the agreement expressed in a backchannel. In evaluations on the MultiMediate'22 backchannel detection challenge, we reach 66.4% accuracy with a one-layer transformer architecture, outperforming the previous state of the art. With a two-layer transformer architecture, we furthermore set a new state of the art (0.0604 MSE) on the task of estimating the amount of agreement expressed in a backchannel.
comment: Accepted at IEEE IJCNN'23
☆ GANs Settle Scores!
Generative adversarial networks (GANs) comprise a generator, trained to learn the underlying distribution of the desired data, and a discriminator, trained to distinguish real samples from those output by the generator. A majority of GAN literature focuses on understanding the optimality of the discriminator through integral probability metric (IPM) or divergence based analysis. In this paper, we propose a unified approach to analyzing the generator optimization through variational approach. In $f$-divergence-minimizing GANs, we show that the optimal generator is the one that matches the score of its output distribution with that of the data distribution, while in IPM GANs, we show that this optimal generator matches score-like functions, involving the flow-field of the kernel associated with a chosen IPM constraint space. Further, the IPM-GAN optimization can be seen as one of smoothed score-matching, where the scores of the data and the generator distributions are convolved with the kernel associated with the constraint. The proposed approach serves to unify score-based training and existing GAN flavors, leveraging results from normalizing flows, while also providing explanations for empirical phenomena such as the stability of non-saturating GAN losses. Based on these results, we propose novel alternatives to $f$-GAN and IPM-GAN training based on score and flow matching, and discriminator-guided Langevin sampling.
☆ Automatic Reconstruction of Semantic 3D Models from 2D Floor Plans
Digitalization of existing buildings and the creation of 3D BIM models for them has become crucial for many tasks. Of particular importance are floor plans, which contain information about building layouts and are vital for processes such as construction, maintenance or refurbishing. However, this data is not always available in digital form, especially for older buildings constructed before CAD tools were widely available, or lacks semantic information. The digitalization of such information usually requires manual work of an expert that must reconstruct the layouts by hand, which is a cumbersome and error-prone process. In this paper, we present a pipeline for reconstruction of vectorized 3D models from scanned 2D plans, aiming at increasing the efficiency of this process. The method presented achieves state-of-the-art results in the public dataset CubiCasa5k, and shows good generalization to different types of plans. Our vectorization approach is particularly effective, outperforming previous methods.
comment: 5 pages, 1 figure
☆ A Conditional Normalizing Flow for Accelerated Multi-Coil MR Imaging ICML 2023
Accelerated magnetic resonance (MR) imaging attempts to reduce acquisition time by collecting data below the Nyquist rate. As an ill-posed inverse problem, many plausible solutions exist, yet the majority of deep learning approaches generate only a single solution. We instead focus on sampling from the posterior distribution, which provides more comprehensive information for downstream inference tasks. To do this, we design a novel conditional normalizing flow (CNF) that infers the signal component in the measurement operator's nullspace, which is later combined with measured data to form complete images. Using fastMRI brain and knee data, we demonstrate fast inference and accuracy that surpasses recent posterior sampling techniques for MRI. Code is available at https://github.com/jwen307/mri_cnf/
comment: Accepted to ICML 2023
☆ HomE: Homography-Equivariant Video Representation Learning
Recent advances in self-supervised representation learning have enabled more efficient and robust model performance without relying on extensive labeled data. However, most works are still focused on images, with few working on videos and even fewer on multi-view videos, where more powerful inductive biases can be leveraged for self-supervision. In this work, we propose a novel method for representation learning of multi-view videos, where we explicitly model the representation space to maintain Homography Equivariance (HomE). Our method learns an implicit mapping between different views, culminating in a representation space that maintains the homography relationship between neighboring views. We evaluate our HomE representation via action recognition and pedestrian intent prediction as downstream tasks. On action classification, our method obtains 96.4% 3-fold accuracy on the UCF101 dataset, better than most state-of-the-art self-supervised learning methods. Similarly, on the STIP dataset, we outperform the state-of-the-art by 6% for pedestrian intent prediction one second into the future while also obtaining an accuracy of 91.2% for pedestrian action (cross vs. not-cross) classification. Code is available at https://github.com/anirudhs123/HomE.
comment: 10 pages, 4 figures, 4 tables
☆ Towards Source-free Domain Adaptive Semantic Segmentation via Importance-aware and Prototype-contrast Learning
Domain adaptive semantic segmentation enables robust pixel-wise understanding in real-world driving scenes. Source-free domain adaptation, as a more practical technique, addresses the concerns of data privacy and storage limitations in typical unsupervised domain adaptation methods. It utilizes a well-trained source model and unlabeled target data to achieve adaptation in the target domain. However, in the absence of source data and target labels, current solutions cannot sufficiently reduce the impact of domain shift and fully leverage the information from the target data. In this paper, we propose an end-to-end source-free domain adaptation semantic segmentation method via Importance-Aware and Prototype-Contrast (IAPC) learning. The proposed IAPC framework effectively extracts domain-invariant knowledge from the well-trained source model and learns domain-specific knowledge from the unlabeled target domain. Specifically, considering the problem of domain shift in the prediction of the target domain by the source model, we put forward an importance-aware mechanism for the biased target prediction probability distribution to extract domain-invariant knowledge from the source model. We further introduce a prototype-contrast strategy, which includes a prototype-symmetric cross-entropy loss and a prototype-enhanced cross-entropy loss, to learn target intra-domain knowledge without relying on labels. A comprehensive variety of experiments on two domain adaptive semantic segmentation benchmarks demonstrates that the proposed end-to-end IAPC solution outperforms existing state-of-the-art methods. Code will be made publicly available at https://github.com/yihong-97/Source-free_IAPC.
comment: Code will be made publicly available at https://github.com/yihong-97/Source-free_IAPC
☆ Two-View Geometry Scoring Without Correspondences
Camera pose estimation for two-view geometry traditionally relies on RANSAC. Normally, a multitude of image correspondences leads to a pool of proposed hypotheses, which are then scored to find a winning model. The inlier count is generally regarded as a reliable indicator of "consensus". We examine this scoring heuristic, and find that it favors disappointing models under certain circumstances. As a remedy, we propose the Fundamental Scoring Network (FSNet), which infers a score for a pair of overlapping images and any proposed fundamental matrix. It does not rely on sparse correspondences, but rather embodies a two-view geometry model through an epipolar attention mechanism that predicts the pose error of the two images. FSNet can be incorporated into traditional RANSAC loops. We evaluate FSNet on fundamental and essential matrix estimation on indoor and outdoor datasets, and establish that FSNet can successfully identify good poses for pairs of images with few or unreliable correspondences. Besides, we show that naively combining FSNet with MAGSAC++ scoring approach achieves state of the art results.
☆ A Novel Vision Transformer with Residual in Self-attention for Biomedical Image Classification
Biomedical image classification requires capturing of bio-informatics based on specific feature distribution. In most of such applications, there are mainly challenges due to limited availability of samples for diseased cases and imbalanced nature of dataset. This article presents the novel framework of multi-head self-attention for vision transformer (ViT) which makes capable of capturing the specific image features for classification and analysis. The proposed method uses the concept of residual connection for accumulating the best attention output in each block of multi-head attention. The proposed framework has been evaluated on two small datasets: (i) blood cell classification dataset and (ii) brain tumor detection using brain MRI images. The results show the significant improvement over traditional ViT and other convolution based state-of-the-art classification models.
☆ Probabilistic Concept Bottleneck Models ICML
Interpretable models are designed to make decisions in a human-interpretable manner. Representatively, Concept Bottleneck Models (CBM) follow a two-step process of concept prediction and class prediction based on the predicted concepts. CBM provides explanations with high-level concepts derived from concept predictions; thus, reliable concept predictions are important for trustworthiness. In this study, we address the ambiguity issue that can harm reliability. While the existence of a concept can often be ambiguous in the data, CBM predicts concepts deterministically without considering this ambiguity. To provide a reliable interpretation against this ambiguity, we propose Probabilistic Concept Bottleneck Models (ProbCBM). By leveraging probabilistic concept embeddings, ProbCBM models uncertainty in concept prediction and provides explanations based on the concept and its corresponding uncertainty. This uncertainty enhances the reliability of the explanations. Furthermore, as class uncertainty is derived from concept uncertainty in ProbCBM, we can explain class uncertainty by means of concept uncertainty. Code is publicly available at https://github.com/ejkim47/prob-cbm.
comment: International Conference on Machine Learning (ICML) 2023
☆ Segment Anything in High Quality
The recent Segment Anything Model (SAM) represents a big leap in scaling up segmentation models, allowing for powerful zero-shot capabilities and flexible prompting. Despite being trained with 1.1 billion masks, SAM's mask prediction quality falls short in many cases, particularly when dealing with objects that have intricate structures. We propose HQ-SAM, equipping SAM with the ability to accurately segment any object, while maintaining SAM's original promptable design, efficiency, and zero-shot generalizability. Our careful design reuses and preserves the pre-trained model weights of SAM, while only introducing minimal additional parameters and computation. We design a learnable High-Quality Output Token, which is injected into SAM's mask decoder and is responsible for predicting the high-quality mask. Instead of only applying it on mask-decoder features, we first fuse them with early and final ViT features for improved mask details. To train our introduced learnable parameters, we compose a dataset of 44K fine-grained masks from several sources. HQ-SAM is only trained on the introduced detaset of 44k masks, which takes only 4 hours on 8 GPUs. We show the efficacy of HQ-SAM in a suite of 9 diverse segmentation datasets across different downstream tasks, where 7 out of them are evaluated in a zero-shot transfer protocol. Our code and models will be released at https://github.com/SysCV/SAM-HQ.
comment: We propose HQ-SAM to upgrade SAM for high-quality zero-shot segmentation. Github: https://github.com/SysCV/SAM-HQ
☆ An Attentive-based Generative Model for Medical Image Synthesis
Magnetic resonance (MR) and computer tomography (CT) imaging are valuable tools for diagnosing diseases and planning treatment. However, limitations such as radiation exposure and cost can restrict access to certain imaging modalities. To address this issue, medical image synthesis can generate one modality from another, but many existing models struggle with high-quality image synthesis when multiple slices are present in the dataset. This study proposes an attention-based dual contrast generative model, called ADC-cycleGAN, which can synthesize medical images from unpaired data with multiple slices. The model integrates a dual contrast loss term with the CycleGAN loss to ensure that the synthesized images are distinguishable from the source domain. Additionally, an attention mechanism is incorporated into the generators to extract informative features from both channel and spatial domains. To improve performance when dealing with multiple slices, the $K$-means algorithm is used to cluster the dataset into $K$ groups, and each group is used to train a separate ADC-cycleGAN. Experimental results demonstrate that the proposed ADC-cycleGAN model produces comparable samples to other state-of-the-art generative models, achieving the highest PSNR and SSIM values of 19.04385 and 0.68551, respectively. We publish the code at https://github.com/JiayuanWang-JW/ADC-cycleGAN.
☆ Publicly available datasets of breast histopathology H&E whole-slide images: A systematic review
Advancements in digital pathology and computing resources have made a significant impact in the field of computational pathology for breast cancer diagnosis and treatment. However, access to high-quality labeled histopathological images of breast cancer is a big challenge that limits the development of accurate and robust deep learning models. In this systematic review, we identified the publicly available datasets of breast H&E stained whole-slide images (WSI) that can be used to develop deep learning algorithms. We systematically searched nine scientific literature databases and nine research data repositories. We found twelve publicly available datasets, containing 5153 H&E WSIs of breast cancer. Moreover, we reported image metadata and characteristics for each dataset to assist researchers in selecting proper datasets for specific tasks in breast cancer computational pathology. In addition, we compiled a list of patch and private datasets that were used in the included articles as a supplementary resource for researchers. Notably, 22% of the included articles utilized multiple datasets, and only 12% of the articles used an external validation set, suggesting that the performance of other developed models may be susceptible to overestimation. The TCGA-BRCA was used in 47.4% of the selected studies. This dataset has a considerable selection bias that can impact the robustness and generalizability of the trained algorithms. There is also a lack of consistent metadata reporting of breast WSI datasets that can be an issue in developing accurate deep learning models, indicating the necessity of establishing explicit guidelines for documenting breast WSI dataset characteristics and metadata.
comment: 27 pages (including references), 6 figures, 3 tables, 4 supporting information materials
☆ PanoGRF: Generalizable Spherical Radiance Fields for Wide-baseline Panoramas
Achieving an immersive experience enabling users to explore virtual environments with six degrees of freedom (6DoF) is essential for various applications such as virtual reality (VR). Wide-baseline panoramas are commonly used in these applications to reduce network bandwidth and storage requirements. However, synthesizing novel views from these panoramas remains a key challenge. Although existing neural radiance field methods can produce photorealistic views under narrow-baseline and dense image captures, they tend to overfit the training views when dealing with \emph{wide-baseline} panoramas due to the difficulty in learning accurate geometry from sparse $360^{\circ}$ views. To address this problem, we propose PanoGRF, Generalizable Spherical Radiance Fields for Wide-baseline Panoramas, which construct spherical radiance fields incorporating $360^{\circ}$ scene priors. Unlike generalizable radiance fields trained on perspective images, PanoGRF avoids the information loss from panorama-to-perspective conversion and directly aggregates geometry and appearance features of 3D sample points from each panoramic view based on spherical projection. Moreover, as some regions of the panorama are only visible from one view while invisible from others under wide baseline settings, PanoGRF incorporates $360^{\circ}$ monocular depth priors into spherical depth estimation to improve the geometry features. Experimental results on multiple panoramic datasets demonstrate that PanoGRF significantly outperforms state-of-the-art generalizable view synthesis methods for wide-baseline panoramas (e.g., OmniSyn) and perspective images (e.g., IBRNet, NeuRay).
☆ Group channel pruning and spatial attention distilling for object detection
Due to the over-parameterization of neural networks, many model compression methods based on pruning and quantization have emerged. They are remarkable in reducing the size, parameter number, and computational complexity of the model. However, most of the models compressed by such methods need the support of special hardware and software, which increases the deployment cost. Moreover, these methods are mainly used in classification tasks, and rarely directly used in detection tasks. To address these issues, for the object detection network we introduce a three-stage model compression method: dynamic sparse training, group channel pruning, and spatial attention distilling. Firstly, to select out the unimportant channels in the network and maintain a good balance between sparsity and accuracy, we put forward a dynamic sparse training method, which introduces a variable sparse rate, and the sparse rate will change with the training process of the network. Secondly, to reduce the effect of pruning on network accuracy, we propose a novel pruning method called group channel pruning. In particular, we divide the network into multiple groups according to the scales of the feature layer and the similarity of module structure in the network, and then we use different pruning thresholds to prune the channels in each group. Finally, to recover the accuracy of the pruned network, we use an improved knowledge distillation method for the pruned network. Especially, we extract spatial attention information from the feature maps of specific scales in each group as knowledge for distillation. In the experiments, we use YOLOv4 as the object detection network and PASCAL VOC as the training dataset. Our method reduces the parameters of the model by 64.7 % and the calculation by 34.9%.
comment: Appl Intell
Transformer-based Multi-Modal Learning for Multi Label Remote Sensing Image Classification
In this paper, we introduce a novel Synchronized Class Token Fusion (SCT Fusion) architecture in the framework of multi-modal multi-label classification (MLC) of remote sensing (RS) images. The proposed architecture leverages modality-specific attention-based transformer encoders to process varying input modalities, while exchanging information across modalities by synchronizing the special class tokens after each transformer encoder block. The synchronization involves fusing the class tokens with a trainable fusion transformation, resulting in a synchronized class token that contains information from all modalities. As the fusion transformation is trainable, it allows to reach an accurate representation of the shared features among different modalities. Experimental results show the effectiveness of the proposed architecture over single-modality architectures and an early fusion multi-modal architecture when evaluated on a multi-modal MLC dataset. The code of the proposed architecture is publicly available at https://git.tu-berlin.de/rsim/sct-fusion.
comment: Accepted at IEEE International Geoscience and Remote Sensing Symposium 2023
☆ A Feature Reuse Framework with Texture-adaptive Aggregation for Reference-based Super-Resolution
Reference-based super-resolution (RefSR) has gained considerable success in the field of super-resolution with the addition of high-resolution reference images to reconstruct low-resolution (LR) inputs with more high-frequency details, thereby overcoming some limitations of single image super-resolution (SISR). Previous research in the field of RefSR has mostly focused on two crucial aspects. The first is accurate correspondence matching between the LR and the reference (Ref) image. The second is the effective transfer and aggregation of similar texture information from the Ref images. Nonetheless, an important detail of perceptual loss and adversarial loss has been underestimated, which has a certain adverse effect on texture transfer and reconstruction. In this study, we propose a feature reuse framework that guides the step-by-step texture reconstruction process through different stages, reducing the negative impacts of perceptual and adversarial loss. The feature reuse framework can be used for any RefSR model, and several RefSR approaches have improved their performance after being retrained using our framework. Additionally, we introduce a single image feature embedding module and a texture-adaptive aggregation module. The single image feature embedding module assists in reconstructing the features of the LR inputs itself and effectively lowers the possibility of including irrelevant textures. The texture-adaptive aggregation module dynamically perceives and aggregates texture information between the LR inputs and the Ref images using dynamic filters. This enhances the utilization of the reference texture while reducing reference misuse. The source code is available at https://github.com/Yi-Yang355/FRFSR.
☆ PolyDiffuse: Polygonal Shape Reconstruction via Guided Set Diffusion Models
This paper presents PolyDiffuse, a novel structured reconstruction algorithm that transforms visual sensor data into polygonal shapes with Diffusion Models (DM), an emerging machinery amid exploding generative AI, while formulating reconstruction as a generation process conditioned on sensor data. The task of structured reconstruction poses two fundamental challenges to DM: 1) A structured geometry is a ``set'' (e.g., a set of polygons for a floorplan geometry), where a sample of $N$ elements has $N!$ different but equivalent representations, making the denoising highly ambiguous; and 2) A ``reconstruction'' task has a single solution, where an initial noise needs to be chosen carefully, while any initial noise works for a generation task. Our technical contribution is the introduction of a Guided Set Diffusion Model where 1) the forward diffusion process learns guidance networks to control noise injection so that one representation of a sample remains distinct from its other permutation variants, thus resolving denoising ambiguity; and 2) the reverse denoising process reconstructs polygonal shapes, initialized and directed by the guidance networks, as a conditional generation process subject to the sensor data. We have evaluated our approach for reconstructing two types of polygonal shapes: floorplan as a set of polygons and HD map for autonomous cars as a set of polylines. Through extensive experiments on standard benchmarks, we demonstrate that PolyDiffuse significantly advances the current state of the art and enables broader practical applications.
comment: Project page: https://poly-diffuse.github.io/
☆ dugMatting: Decomposed-Uncertainty-Guided Matting
Cutting out an object and estimating its opacity mask, known as image matting, is a key task in image and video editing. Due to the highly ill-posed issue, additional inputs, typically user-defined trimaps or scribbles, are usually needed to reduce the uncertainty. Although effective, it is either time consuming or only suitable for experienced users who know where to place the strokes. In this work, we propose a decomposed-uncertainty-guided matting (dugMatting) algorithm, which explores the explicitly decomposed uncertainties to efficiently and effectively improve the results. Basing on the characteristic of these uncertainties, the epistemic uncertainty is reduced in the process of guiding interaction (which introduces prior knowledge), while the aleatoric uncertainty is reduced in modeling data distribution (which introduces statistics for both data and possible noise). The proposed matting framework relieves the requirement for users to determine the interaction areas by using simple and efficient labeling. Extensively quantitative and qualitative results validate that the proposed method significantly improves the original matting algorithms in terms of both efficiency and efficacy.
☆ SASMU: boost the performance of generalized recognition model using synthetic face dataset
Nowadays, deploying a robust face recognition product becomes easy with the development of face recognition techniques for decades. Not only profile image verification but also the state-of-the-art method can handle the in-the-wild image almost perfectly. However, the concern of privacy issues raise rapidly since mainstream research results are powered by tons of web-crawled data, which faces the privacy invasion issue. The community tries to escape this predicament completely by training the face recognition model with synthetic data but faces severe domain gap issues, which still need to access real images and identity labels to fine-tune the model. In this paper, we propose SASMU, a simple, novel, and effective method for face recognition using a synthetic dataset. Our proposed method consists of spatial data augmentation (SA) and spectrum mixup (SMU). We first analyze the existing synthetic datasets for developing a face recognition system. Then, we reveal that heavy data augmentation is helpful for boosting performance when using synthetic data. By analyzing the previous frequency mixup studies, we proposed a novel method for domain generalization. Extensive experimental results have demonstrated the effectiveness of SASMU, achieving state-of-the-art performance on several common benchmarks, such as LFW, AgeDB-30, CA-LFW, CFP-FP, and CP-LFW.
comment: under review
☆ Bi-LRFusion: Bi-Directional LiDAR-Radar Fusion for 3D Dynamic Object Detection CVPR2023
LiDAR and Radar are two complementary sensing approaches in that LiDAR specializes in capturing an object's 3D shape while Radar provides longer detection ranges as well as velocity hints. Though seemingly natural, how to efficiently combine them for improved feature representation is still unclear. The main challenge arises from that Radar data are extremely sparse and lack height information. Therefore, directly integrating Radar features into LiDAR-centric detection networks is not optimal. In this work, we introduce a bi-directional LiDAR-Radar fusion framework, termed Bi-LRFusion, to tackle the challenges and improve 3D detection for dynamic objects. Technically, Bi-LRFusion involves two steps: first, it enriches Radar's local features by learning important details from the LiDAR branch to alleviate the problems caused by the absence of height information and extreme sparsity; second, it combines LiDAR features with the enhanced Radar features in a unified bird's-eye-view representation. We conduct extensive experiments on nuScenes and ORR datasets, and show that our Bi-LRFusion achieves state-of-the-art performance for detecting dynamic objects. Notably, Radar data in these two datasets have different formats, which demonstrates the generalizability of our method. Codes are available at https://github.com/JessieW0806/BiLRFusion.
comment: accepted by CVPR2023
☆ Leveraging the Triple Exponential Moving Average for Fast-Adaptive Moment Estimation
Network optimization is a crucial step in the field of deep learning, as it directly affects the performance of models in various domains such as computer vision. Despite the numerous optimizers that have been developed over the years, the current methods are still limited in their ability to accurately and quickly identify gradient trends, which can lead to sub-optimal network performance. In this paper, we propose a novel deep optimizer called Fast-Adaptive Moment Estimation (FAME), which for the first time estimates gradient moments using a Triple Exponential Moving Average (TEMA). Incorporating TEMA into the optimization process provides richer and more accurate information on data changes and trends, as compared to the standard Exponential Moving Average used in essentially all current leading adaptive optimization methods. Our proposed FAME optimizer has been extensively validated through a wide range of benchmarks, including CIFAR-10, CIFAR-100, PASCAL-VOC, MS-COCO, and Cityscapes, using 14 different learning architectures, six optimizers, and various vision tasks, including detection, classification and semantic understanding. The results demonstrate that our FAME optimizer outperforms other leading optimizers in terms of both robustness and accuracy.
☆ Convergence analysis of equilibrium methods for inverse problems
Recently, the use of deep equilibrium methods has emerged as a new approach for solving imaging and other ill-posed inverse problems. While learned components may be a key factor in the good performance of these methods in practice, a theoretical justification from a regularization point of view is still lacking. In this paper, we address this issue by providing stability and convergence results for the class of equilibrium methods. In addition, we derive convergence rates and stability estimates in the symmetric Bregman distance. We strengthen our results for regularization operators with contractive residuals. Furthermore, we use the presented analysis to gain insight into the practical behavior of these methods, including a lower bound on the performance of the regularized solutions. In addition, we show that the convergence analysis leads to the design of a new type of loss function which has several advantages over previous ones. Numerical simulations are used to support our findings.
☆ Learning Landmarks Motion from Speech for Speaker-Agnostic 3D Talking Heads Generation
This paper presents a novel approach for generating 3D talking heads from raw audio inputs. Our method grounds on the idea that speech related movements can be comprehensively and efficiently described by the motion of a few control points located on the movable parts of the face, i.e., landmarks. The underlying musculoskeletal structure then allows us to learn how their motion influences the geometrical deformations of the whole face. The proposed method employs two distinct models to this aim: the first one learns to generate the motion of a sparse set of landmarks from the given audio. The second model expands such landmarks motion to a dense motion field, which is utilized to animate a given 3D mesh in neutral state. Additionally, we introduce a novel loss function, named Cosine Loss, which minimizes the angle between the generated motion vectors and the ground truth ones. Using landmarks in 3D talking head generation offers various advantages such as consistency, reliability, and obviating the need for manual-annotation. Our approach is designed to be identity-agnostic, enabling high-quality facial animations for any users without additional data or training.
☆ Learning Signed Distance Functions from Noisy 3D Point Clouds via Noise to Noise Mapping ICML2023
Learning signed distance functions (SDFs) from 3D point clouds is an important task in 3D computer vision. However, without ground truth signed distances, point normals or clean point clouds, current methods still struggle from learning SDFs from noisy point clouds. To overcome this challenge, we propose to learn SDFs via a noise to noise mapping, which does not require any clean point cloud or ground truth supervision for training. Our novelty lies in the noise to noise mapping which can infer a highly accurate SDF of a single object or scene from its multiple or even single noisy point cloud observations. Our novel learning manner is supported by modern Lidar systems which capture multiple noisy observations per second. We achieve this by a novel loss which enables statistical reasoning on point clouds and maintains geometric consistency although point clouds are irregular, unordered and have no point correspondence among noisy observations. Our evaluation under the widely used benchmarks demonstrates our superiority over the state-of-the-art methods in surface reconstruction, point cloud denoising and upsampling. Our code, data, and pre-trained models are available at https://github.com/mabaorui/Noise2NoiseMapping/
comment: To appear at ICML2023. Code and data are available at https://github.com/mabaorui/Noise2NoiseMapping/
☆ Evaluating The Robustness of Self-Supervised Representations to Background/Foreground Removal
Despite impressive empirical advances of SSL in solving various tasks, the problem of understanding and characterizing SSL representations learned from input data remains relatively under-explored. We provide a comparative analysis of how the representations produced by SSL models differ when masking parts of the input. Specifically, we considered state-of-the-art SSL pretrained models, such as DINOv2, MAE, and SwaV, and analyzed changes at the representation levels across 4 Image Classification datasets. First, we generate variations of the datasets by applying foreground and background segmentation. Then, we conduct statistical analysis using Canonical Correlation Analysis (CCA) and Centered Kernel Alignment (CKA) to evaluate the robustness of the representations learned in SSL models. Empirically, we show that not all models lead to representations that separate foreground, background, and complete images. Furthermore, we test different masking strategies by occluding the center regions of the images to address cases where foreground and background are difficult. For example, the DTD dataset that focuses on texture rather specific objects.
☆ Masked Autoencoder for Unsupervised Video Summarization
Summarizing a video requires a diverse understanding of the video, ranging from recognizing scenes to evaluating how much each frame is essential enough to be selected as a summary. Self-supervised learning (SSL) is acknowledged for its robustness and flexibility to multiple downstream tasks, but the video SSL has not shown its value for dense understanding tasks like video summarization. We claim an unsupervised autoencoder with sufficient self-supervised learning does not need any extra downstream architecture design or fine-tuning weights to be utilized as a video summarization model. The proposed method to evaluate the importance score of each frame takes advantage of the reconstruction score of the autoencoder's decoder. We evaluate the method in major unsupervised video summarization benchmarks to show its effectiveness under various experimental settings.
☆ Robust and Generalisable Segmentation of Subtle Epilepsy-causing Lesions: a Graph Convolutional Approach MICCAI 2023
Focal cortical dysplasia (FCD) is a leading cause of drug-resistant focal epilepsy, which can be cured by surgery. These lesions are extremely subtle and often missed even by expert neuroradiologists. "Ground truth" manual lesion masks are therefore expensive, limited and have large inter-rater variability. Existing FCD detection methods are limited by high numbers of false positive predictions, primarily due to vertex- or patch-based approaches that lack whole-brain context. Here, we propose to approach the problem as semantic segmentation using graph convolutional networks (GCN), which allows our model to learn spatial relationships between brain regions. To address the specific challenges of FCD identification, our proposed model includes an auxiliary loss to predict distance from the lesion to reduce false positives and a weak supervision classification loss to facilitate learning from uncertain lesion masks. On a multi-centre dataset of 1015 participants with surface-based features and manual lesion masks from structural MRI data, the proposed GCN achieved an AUC of 0.74, a significant improvement against a previously used vertex-wise multi-layer perceptron (MLP) classifier (AUC 0.64). With sensitivity thresholded at 67%, the GCN had a specificity of 71% in comparison to 49% when using the MLP. This improvement in specificity is vital for clinical integration of lesion-detection tools into the radiological workflow, through increasing clinical confidence in the use of AI radiological adjuncts and reducing the number of areas requiring expert review.
comment: accepted at MICCAI 2023
☆ Towards Robust GAN-generated Image Detection: a Multi-view Completion Representation IJCAI 2023
GAN-generated image detection now becomes the first line of defense against the malicious uses of machine-synthesized image manipulations such as deepfakes. Although some existing detectors work well in detecting clean, known GAN samples, their success is largely attributable to overfitting unstable features such as frequency artifacts, which will cause failures when facing unknown GANs or perturbation attacks. To overcome the issue, we propose a robust detection framework based on a novel multi-view image completion representation. The framework first learns various view-to-image tasks to model the diverse distributions of genuine images. Frequency-irrelevant features can be represented from the distributional discrepancies characterized by the completion models, which are stable, generalized, and robust for detecting unknown fake patterns. Then, a multi-view classification is devised with elaborated intra- and inter-view learning strategies to enhance view-specific feature representation and cross-view feature aggregation, respectively. We evaluated the generalization ability of our framework across six popular GANs at different resolutions and its robustness against a broad range of perturbation attacks. The results confirm our method's improved effectiveness, generalization, and robustness over various baselines.
comment: Accepted to IJCAI 2023
☆ Quantifying Sample Anonymity in Score-Based Generative Models with Adversarial Fingerprinting
Recent advances in score-based generative models have led to a huge spike in the development of downstream applications using generative models ranging from data augmentation over image and video generation to anomaly detection. Despite publicly available trained models, their potential to be used for privacy preserving data sharing has not been fully explored yet. Training diffusion models on private data and disseminating the models and weights rather than the raw dataset paves the way for innovative large-scale data-sharing strategies, particularly in healthcare, where safeguarding patients' personal health information is paramount. However, publishing such models without individual consent of, e.g., the patients from whom the data was acquired, necessitates guarantees that identifiable training samples will never be reproduced, thus protecting personal health data and satisfying the requirements of policymakers and regulatory bodies. This paper introduces a method for estimating the upper bound of the probability of reproducing identifiable training images during the sampling process. This is achieved by designing an adversarial approach that searches for anatomic fingerprints, such as medical devices or dermal art, which could potentially be employed to re-identify training images. Our method harnesses the learned score-based model to estimate the probability of the entire subspace of the score function that may be utilized for one-to-one reproduction of training samples. To validate our estimates, we generate anomalies containing a fingerprint and investigate whether generated samples from trained generative models can be uniquely mapped to the original training samples. Overall our results show that privacy-breaching images are reproduced at sampling time if the models were trained without care.
comment: 10 pages, 6 figures
☆ DWT-CompCNN: Deep Image Classification Network for High Throughput JPEG 2000 Compressed Documents
For any digital application with document images such as retrieval, the classification of document images becomes an essential stage. Conventionally for the purpose, the full versions of the documents, that is the uncompressed document images make the input dataset, which poses a threat due to the big volume required to accommodate the full versions of the documents. Therefore, it would be novel, if the same classification task could be accomplished directly (with some partial decompression) with the compressed representation of documents in order to make the whole process computationally more efficient. In this research work, a novel deep learning model, DWT CompCNN is proposed for classification of documents that are compressed using High Throughput JPEG 2000 (HTJ2K) algorithm. The proposed DWT-CompCNN comprises of five convolutional layers with filter sizes of 16, 32, 64, 128, and 256 consecutively for each increasing layer to improve learning from the wavelet coefficients extracted from the compressed images. Experiments are performed on two benchmark datasets- Tobacco-3482 and RVL-CDIP, which demonstrate that the proposed model is time and space efficient, and also achieves a better classification accuracy in compressed domain.
comment: In Springer Journal - Pattern Analysis and Applications under Minor Revision
☆ Adjustable Visual Appearance for Generalizable Novel View Synthesis
We present a generalizable novel view synthesis method where it is possible to modify the visual appearance of rendered views to match a target weather or lighting condition. Our method is based on a generalizable transformer architecture, trained on synthetically generated scenes under different appearance conditions. This allows for rendering novel views in a consistent manner of 3D scenes that were not included in the training set, along with the ability to (i) modify their appearance to match the target condition and (ii) smoothly interpolate between different conditions. Experiments on both real and synthetic scenes are provided including both qualitative and quantitative evaluations. Please refer to our project page for video results: https://ava-nvs.github.io/
☆ Bilevel Fast Scene Adaptation for Low-Light Image Enhancement
Enhancing images in low-light scenes is a challenging but widely concerned task in the computer vision. The mainstream learning-based methods mainly acquire the enhanced model by learning the data distribution from the specific scenes, causing poor adaptability (even failure) when meeting real-world scenarios that have never been encountered before. The main obstacle lies in the modeling conundrum from distribution discrepancy across different scenes. To remedy this, we first explore relationships between diverse low-light scenes based on statistical analysis, i.e., the network parameters of the encoder trained in different data distributions are close. We introduce the bilevel paradigm to model the above latent correspondence from the perspective of hyperparameter optimization. A bilevel learning framework is constructed to endow the scene-irrelevant generality of the encoder towards diverse scenes (i.e., freezing the encoder in the adaptation and testing phases). Further, we define a reinforced bilevel learning framework to provide a meta-initialization for scene-specific decoder to further ameliorate visual quality. Moreover, to improve the practicability, we establish a Retinex-induced architecture with adaptive denoising and apply our built learning framework to acquire its parameters by using two training losses including supervised and unsupervised forms. Extensive experimental evaluations on multiple datasets verify our adaptability and competitive performance against existing state-of-the-art works. The code and datasets will be available at https://github.com/vis-opt-group/BL.
Transformer-based Annotation Bias-aware Medical Image Segmentation
Manual medical image segmentation is subjective and suffers from annotator-related bias, which can be mimicked or amplified by deep learning methods. Recently, researchers have suggested that such bias is the combination of the annotator preference and stochastic error, which are modeled by convolution blocks located after decoder and pixel-wise independent Gaussian distribution, respectively. It is unlikely that convolution blocks can effectively model the varying degrees of preference at the full resolution level. Additionally, the independent pixel-wise Gaussian distribution disregards pixel correlations, leading to a discontinuous boundary. This paper proposes a Transformer-based Annotation Bias-aware (TAB) medical image segmentation model, which tackles the annotator-related bias via modeling annotator preference and stochastic errors. TAB employs the Transformer with learnable queries to extract the different preference-focused features. This enables TAB to produce segmentation with various preferences simultaneously using a single segmentation head. Moreover, TAB takes the multivariant normal distribution assumption that models pixel correlations, and learns the annotation distribution to disentangle the stochastic error. We evaluated our TAB on an OD/OC segmentation benchmark annotated by six annotators. Our results suggest that TAB outperforms existing medical image segmentation models which take into account the annotator-related bias.
comment: 11 pages, 2 figures
☆ Privacy Distillation: Reducing Re-identification Risk of Multimodal Diffusion Models
Knowledge distillation in neural networks refers to compressing a large model or dataset into a smaller version of itself. We introduce Privacy Distillation, a framework that allows a text-to-image generative model to teach another model without exposing it to identifiable data. Here, we are interested in the privacy issue faced by a data provider who wishes to share their data via a multimodal generative model. A question that immediately arises is ``How can a data provider ensure that the generative model is not leaking identifiable information about a patient?''. Our solution consists of (1) training a first diffusion model on real data (2) generating a synthetic dataset using this model and filtering it to exclude images with a re-identifiability risk (3) training a second diffusion model on the filtered synthetic data only. We showcase that datasets sampled from models trained with privacy distillation can effectively reduce re-identification risk whilst maintaining downstream performance.
☆ Independent Modular Networks ICRA23
Monolithic neural networks that make use of a single set of weights to learn useful representations for downstream tasks explicitly dismiss the compositional nature of data generation processes. This characteristic exists in data where every instance can be regarded as the combination of an identity concept, such as the shape of an object, combined with modifying concepts, such as orientation, color, and size. The dismissal of compositionality is especially detrimental in robotics, where state estimation relies heavily on the compositional nature of physical mechanisms (e.g., rotations and transformations) to model interactions. To accommodate this data characteristic, modular networks have been proposed. However, a lack of structure in each module's role, and modular network-specific issues such as module collapse have restricted their usability. We propose a modular network architecture that accommodates the mentioned decompositional concept by proposing a unique structure that splits the modules into predetermined roles. Additionally, we provide regularizations that improve the resiliency of the modular network to the problem of module collapse while improving the decomposition accuracy of the model.
comment: ICRA23 RAP4Robots Workshop
☆ Egocentric Planning for Scalable Embodied Task Achievement
Embodied agents face significant challenges when tasked with performing actions in diverse environments, particularly in generalizing across object types and executing suitable actions to accomplish tasks. Furthermore, agents should exhibit robustness, minimizing the execution of illegal actions. In this work, we present Egocentric Planning, an innovative approach that combines symbolic planning and Object-oriented POMDPs to solve tasks in complex environments, harnessing existing models for visual perception and natural language processing. We evaluated our approach in ALFRED, a simulated environment designed for domestic tasks, and demonstrated its high scalability, achieving an impressive 36.07% unseen success rate in the ALFRED benchmark and winning the ALFRED challenge at CVPR Embodied AI workshop. Our method requires reliable perception and the specification or learning of a symbolic description of the preconditions and effects of the agent's actions, as well as what object types reveal information about others. It is capable of naturally scaling to solve new tasks beyond ALFRED, as long as they can be solved using the available skills. This work offers a solid baseline for studying end-to-end and hybrid methods that aim to generalize to new tasks, including recent approaches relying on LLMs, but often struggle to scale to long sequences of actions or produce robust plans for novel tasks.
☆ LoCoOp: Few-Shot Out-of-Distribution Detection via Prompt Learning
We present a novel vision-language prompt learning approach for few-shot out-of-distribution (OOD) detection. Few-shot OOD detection aims to detect OOD images from classes that are unseen during training using only a few labeled in-distribution (ID) images. While prompt learning methods such as CoOp have shown effectiveness and efficiency in few-shot ID classification, they still face limitations in OOD detection due to the potential presence of ID-irrelevant information in text embeddings. To address this issue, we introduce a new approach called \textbf{Lo}cal regularized \textbf{Co}ntext \textbf{Op}timization (LoCoOp), which performs OOD regularization that utilizes the portions of CLIP local features as OOD features during training. CLIP's local features have a lot of ID-irrelevant nuisances (e.g., backgrounds), and by learning to push them away from the ID class text embeddings, we can remove the nuisances in the ID class text embeddings and enhance the separation between ID and OOD. Experiments on the large-scale ImageNet OOD detection benchmarks demonstrate the superiority of our LoCoOp over zero-shot, fully supervised detection methods and prompt learning methods. Notably, even in a one-shot setting -- just one label per class, LoCoOp outperforms existing zero-shot and fully supervised detection methods. The code will be available via \url{https://github.com/AtsuMiyai/LoCoOp}.
☆ NNMobile-Net: Rethinking CNN Design for Deep Learning-Based Retinopathy Research
Retinal diseases (RD) are the leading cause of severe vision loss or blindness. Deep learning-based automated tools play an indispensable role in assisting clinicians in diagnosing and monitoring RD in modern medicine. Recently, an increasing number of works in this field have taken advantage of Vision Transformer to achieve state-of-the-art performance with more parameters and higher model complexity compared to Convolutional Neural Networks (CNNs). Such sophisticated and task-specific model designs, however, are prone to be overfitting and hinder their generalizability. In this work, we argue that a channel-aware and well-calibrated CNN model may overcome these problems. To this end, we empirically studied CNN's macro and micro designs and its training strategies. Based on the investigation, we proposed a no-new-MobleNet (nn-MobileNet) developed for retinal diseases. In our experiments, our generic, simple and efficient model superseded most current state-of-the-art methods on four public datasets for multiple tasks, including diabetic retinopathy grading, fundus multi-disease detection, and diabetic macular edema classification. Our work may provide novel insights into deep learning architecture design and advance retinopathy research.
comment: Code will publish soon: https://github.com/Retinal-Research/NNMOBILE-NET
☆ DeepfakeArt Challenge: A Benchmark Dataset for Generative AI Art Forgery and Data Poisoning Detection
The tremendous recent advances in generative artificial intelligence techniques have led to significant successes and promise in a wide range of different applications ranging from conversational agents and textual content generation to voice and visual synthesis. Amid the rise in generative AI and its increasing widespread adoption, there has been significant growing concern over the use of generative AI for malicious purposes. In the realm of visual content synthesis using generative AI, key areas of significant concern has been image forgery (e.g., generation of images containing or derived from copyright content), and data poisoning (i.e., generation of adversarially contaminated images). Motivated to address these key concerns to encourage responsible generative AI, we introduce the DeepfakeArt Challenge, a large-scale challenge benchmark dataset designed specifically to aid in the building of machine learning algorithms for generative AI art forgery and data poisoning detection. Comprising of over 32,000 records across a variety of generative forgery and data poisoning techniques, each entry consists of a pair of images that are either forgeries / adversarially contaminated or not. Each of the generated images in the DeepfakeArt Challenge benchmark dataset has been quality checked in a comprehensive manner. The DeepfakeArt Challenge is a core part of GenAI4Good, a global open source initiative for accelerating machine learning for promoting responsible creation and deployment of generative AI for good.
☆ DeepScribe: Localization and Classification of Elamite Cuneiform Signs Via Deep Learning
Twenty-five hundred years ago, the paperwork of the Achaemenid Empire was recorded on clay tablets. In 1933, archaeologists from the University of Chicago's Oriental Institute (OI) found tens of thousands of these tablets and fragments during the excavation of Persepolis. Many of these tablets have been painstakingly photographed and annotated by expert cuneiformists, and now provide a rich dataset consisting of over 5,000 annotated tablet images and 100,000 cuneiform sign bounding boxes. We leverage this dataset to develop DeepScribe, a modular computer vision pipeline capable of localizing cuneiform signs and providing suggestions for the identity of each sign. We investigate the difficulty of learning subtasks relevant to cuneiform tablet transcription on ground-truth data, finding that a RetinaNet object detector can achieve a localization mAP of 0.78 and a ResNet classifier can achieve a top-5 sign classification accuracy of 0.89. The end-to-end pipeline achieves a top-5 classification accuracy of 0.80. As part of the classification module, DeepScribe groups cuneiform signs into morphological clusters. We consider how this automatic clustering approach differs from the organization of standard, printed sign lists and what we may learn from it. These components, trained individually, are sufficient to produce a system that can analyze photos of cuneiform tablets from the Achaemenid period and provide useful transliteration suggestions to researchers. We evaluate the model's end-to-end performance on locating and classifying signs, providing a roadmap to a linguistically-aware transliteration system, then consider the model's potential utility when applied to other periods of cuneiform writing.
comment: Currently under review in the ACM JOCCH
☆ Collect-and-Distribute Transformer for 3D Point Cloud Analysis
Although remarkable advancements have been made recently in point cloud analysis through the exploration of transformer architecture, it remains challenging to effectively learn local and global structures within point clouds. In this paper, we propose a new transformer architecture equipped with a collect-and-distribute mechanism to communicate short- and long-range contexts of point clouds, which we refer to as CDFormer. Specifically, we first utilize self-attention to capture short-range interactions within each local patch, and the updated local features are then collected into a set of proxy reference points from which we can extract long-range contexts. Afterward, we distribute the learned long-range contexts back to local points via cross-attention. To address the position clues for short- and long-range contexts, we also introduce context-aware position encoding to facilitate position-aware communications between points. We perform experiments on four popular point cloud datasets, namely ModelNet40, ScanObjectNN, S3DIS, and ShapeNetPart, for classification and segmentation. Results show the effectiveness of the proposed CDFormer, delivering several new state-of-the-art performances on point cloud classification and segmentation tasks. The code is available at \url{https://github.com/haibo-qiu/CDFormer}.
comment: Code is available at https://github.com/haibo-qiu/CDFormer
☆ Deep Reinforcement Learning Framework for Thoracic Diseases Classification via Prior Knowledge Guidance
The chest X-ray is often utilized for diagnosing common thoracic diseases. In recent years, many approaches have been proposed to handle the problem of automatic diagnosis based on chest X-rays. However, the scarcity of labeled data for related diseases still poses a huge challenge to an accurate diagnosis. In this paper, we focus on the thorax disease diagnostic problem and propose a novel deep reinforcement learning framework, which introduces prior knowledge to direct the learning of diagnostic agents and the model parameters can also be continuously updated as the data increases, like a person's learning process. Especially, 1) prior knowledge can be learned from the pre-trained model based on old data or other domains' similar data, which can effectively reduce the dependence on target domain data, and 2) the framework of reinforcement learning can make the diagnostic agent as exploratory as a human being and improve the accuracy of diagnosis through continuous exploration. The method can also effectively solve the model learning problem in the case of few-shot data and improve the generalization ability of the model. Finally, our approach's performance was demonstrated using the well-known NIH ChestX-ray 14 and CheXpert datasets, and we achieved competitive results. The source code can be found here: \url{https://github.com/NeaseZ/MARL}.
☆ Exploring the Boundaries of Semi-Supervised Facial Expression Recognition: Learning from In-Distribution, Out-of-Distribution, and Unconstrained Data
Deep learning-based methods have been the key driving force behind much of the recent success of facial expression recognition (FER) systems. However, the need for large amounts of labelled data remains a challenge. Semi-supervised learning offers a way to overcome this limitation, allowing models to learn from a small amount of labelled data along with a large unlabelled dataset. While semi-supervised learning has shown promise in FER, most current methods from general computer vision literature have not been explored in the context of FER. In this work, we present a comprehensive study on 11 of the most recent semi-supervised methods, in the context of FER, namely Pi-model, Pseudo-label, Mean Teacher, VAT, UDA, MixMatch, ReMixMatch, FlexMatch, CoMatch, and CCSSL. Our investigation covers semi-supervised learning from in-distribution, out-of-distribution, unconstrained, and very small unlabelled data. Our evaluation includes five FER datasets plus one large face dataset for unconstrained learning. Our results demonstrate that FixMatch consistently achieves better performance on in-distribution unlabelled data, while ReMixMatch stands out among all methods for out-of-distribution, unconstrained, and scarce unlabelled data scenarios. Another significant observation is that semi-supervised learning produces a reasonable improvement over supervised learning, regardless of whether in-distribution, out-of-distribution, or unconstrained data is utilized as the unlabelled set. We also conduct sensitivity analyses on critical hyper-parameters for the two best methods of each setting.
☆ Scaling Up Semi-supervised Learning with Unconstrained Unlabelled Data
We propose UnMixMatch, a semi-supervised learning framework which can learn effective representations from unconstrained unlabelled data in order to scale up performance. Most existing semi-supervised methods rely on the assumption that labelled and unlabelled samples are drawn from the same distribution, which limits the potential for improvement through the use of free-living unlabeled data. Consequently, the generalizability and scalability of semi-supervised learning are often hindered by this assumption. Our method aims to overcome these constraints and effectively utilize unconstrained unlabelled data in semi-supervised learning. UnMixMatch consists of three main components: a supervised learner with hard augmentations that provides strong regularization, a contrastive consistency regularizer to learn underlying representations from the unlabelled data, and a self-supervised loss to enhance the representations that are learnt from the unlabelled data. We perform extensive experiments on 4 commonly used datasets and demonstrate superior performance over existing semi-supervised methods with a performance boost of 4.79%. Extensive ablation and sensitivity studies show the effectiveness and impact of each of the proposed components of our method.
☆ A new method using deep transfer learning on ECG to predict the response to cardiac resynchronization therapy
Background: Cardiac resynchronization therapy (CRT) has emerged as an effective treatment for heart failure patients with electrical dyssynchrony. However, accurately predicting which patients will respond to CRT remains a challenge. This study explores the application of deep transfer learning techniques to train a predictive model for CRT response. Methods: In this study, the short-time Fourier transform (STFT) technique was employed to transform ECG signals into two-dimensional images. A transfer learning approach was then applied on the MIT-BIT ECG database to pre-train a convolutional neural network (CNN) model. The model was fine-tuned to extract relevant features from the ECG images, and then tested on our dataset of CRT patients to predict their response. Results: Seventy-one CRT patients were enrolled in this study. The transfer learning model achieved an accuracy of 72% in distinguishing responders from non-responders in the local dataset. Furthermore, the model showed good sensitivity (0.78) and specificity (0.79) in identifying CRT responders. The performance of our model outperformed clinic guidelines and traditional machine learning approaches. Conclusion: The utilization of ECG images as input and leveraging the power of transfer learning allows for improved accuracy in identifying CRT responders. This approach offers potential for enhancing patient selection and improving outcomes of CRT.
☆ Counting Crowds in Bad Weather
Crowd counting has recently attracted significant attention in the field of computer vision due to its wide applications to image understanding. Numerous methods have been proposed and achieved state-of-the-art performance for real-world tasks. However, existing approaches do not perform well under adverse weather such as haze, rain, and snow since the visual appearances of crowds in such scenes are drastically different from those images in clear weather of typical datasets. In this paper, we propose a method for robust crowd counting in adverse weather scenarios. Instead of using a two-stage approach that involves image restoration and crowd counting modules, our model learns effective features and adaptive queries to account for large appearance variations. With these weather queries, the proposed model can learn the weather information according to the degradation of the input image and optimize with the crowd counting module simultaneously. Experimental results show that the proposed algorithm is effective in counting crowds under different weather types on benchmark datasets. The source code and trained models will be made available to the public.
comment: including supplemental material
♻ ☆ Scaling in Depth: Unlocking Robustness Certification on ImageNet
Despite the promise of Lipschitz-based methods for provably-robust deep learning with deterministic guarantees, current state-of-the-art results are limited to feed-forward Convolutional Networks (ConvNets) on low-dimensional data, such as CIFAR-10. This paper investigates strategies for expanding certifiably robust training to larger, deeper models. A key challenge in certifying deep networks is efficient calculation of the Lipschitz bound for residual blocks found in ResNet and ViT architectures. We show that fast ways of bounding the Lipschitz constant for conventional ResNets are loose, and show how to address this by designing a new residual block, leading to the \emph{Linear ResNet} (LiResNet) architecture. We then introduce \emph{Efficient Margin MAximization} (EMMA), a loss function that stabilizes robust training by simultaneously penalizing worst-case adversarial examples from \emph{all} classes. Together, these contributions yield new \emph{state-of-the-art} robust accuracy on CIFAR-10/100 and Tiny-ImageNet under $\ell_2$ perturbations. Moreover, for the first time, we are able to scale up fast deterministic robustness guarantees to ImageNet, demonstrating that this approach to robust learning can be applied to real-world applications. We release our code on Github: \url{https://github.com/klasleino/gloro}.
♻ ☆ Diffusion Self-Guidance for Controllable Image Generation
Large-scale generative models are capable of producing high-quality images from detailed text descriptions. However, many aspects of an image are difficult or impossible to convey through text. We introduce self-guidance, a method that provides greater control over generated images by guiding the internal representations of diffusion models. We demonstrate that properties such as the shape, location, and appearance of objects can be extracted from these representations and used to steer sampling. Self-guidance works similarly to classifier guidance, but uses signals present in the pretrained model itself, requiring no additional models or training. We show how a simple set of properties can be composed to perform challenging image manipulations, such as modifying the position or size of objects, merging the appearance of objects in one image with the layout of another, composing objects from many images into one, and more. We also show that self-guidance can be used to edit real images. For results and an interactive demo, see our project page at https://dave.ml/selfguidance/
comment: Project page at https://dave.ml/selfguidance/
♻ ☆ InternGPT: Solving Vision-Centric Tasks by Interacting with ChatGPT Beyond Language
We present an interactive visual framework named InternGPT, or iGPT for short. The framework integrates chatbots that have planning and reasoning capabilities, such as ChatGPT, with non-verbal instructions like pointing movements that enable users to directly manipulate images or videos on the screen. Pointing (including gestures, cursors, etc.) movements can provide more flexibility and precision in performing vision-centric tasks that require fine-grained control, editing, and generation of visual content. The name InternGPT stands for \textbf{inter}action, \textbf{n}onverbal, and \textbf{chat}bots. Different from existing interactive systems that rely on pure language, by incorporating pointing instructions, the proposed iGPT significantly improves the efficiency of communication between users and chatbots, as well as the accuracy of chatbots in vision-centric tasks, especially in complicated visual scenarios where the number of objects is greater than 2. Additionally, in iGPT, an auxiliary control mechanism is used to improve the control capability of LLM, and a large vision-language model termed Husky is fine-tuned for high-quality multi-modal dialogue (impressing ChatGPT-3.5-turbo with 93.89\% GPT-4 Quality). We hope this work can spark new ideas and directions for future interactive visual systems. Welcome to watch the code at https://github.com/OpenGVLab/InternGPT.
comment: Technical Report
♻ ☆ Visual Question Answering: A Survey on Techniques and Common Trends in Recent Literature
Visual Question Answering (VQA) is an emerging area of interest for researches, being a recent problem in natural language processing and image prediction. In this area, an algorithm needs to answer questions about certain images. As of the writing of this survey, 25 recent studies were analyzed. Besides, 6 datasets were analyzed and provided their link to download. In this work, several recent pieces of research in this area were investigated and a deeper analysis and comparison among them were provided, including results, the state-of-the-art, common errors, and possible points of improvement for future researchers.
comment: 30 pages. arXiv admin note: text overlap with arXiv:2104.00926, arXiv:2110.02526, arXiv:2108.02059, arXiv:1908.01801 by other authors
♻ ☆ Supervised Metric Learning to Rank for Retrieval via Contextual Similarity Optimization
There is extensive interest in metric learning methods for image retrieval. Many metric learning loss functions focus on learning a correct ranking of training samples, but strongly overfit semantically inconsistent labels and require a large amount of data. To address these shortcomings, we propose a new metric learning method, called contextual loss, which optimizes contextual similarity in addition to cosine similarity. Our contextual loss implicitly enforces semantic consistency among neighbors while converging to the correct ranking. We empirically show that the proposed loss is more robust to label noise, and is less prone to overfitting even when a large portion of train data is withheld. Extensive experiments demonstrate that our method achieves a new state-of-the-art across four image retrieval benchmarks and multiple different evaluation settings. Code is available at: https://github.com/Chris210634/metric-learning-using-contextual-similarity
♻ ☆ Are Deep Neural Networks SMARTer than Second Graders? CVPR 2023
Recent times have witnessed an increasing number of applications of deep neural networks towards solving tasks that require superior cognitive abilities, e.g., playing Go, generating art, ChatGPT, etc. Such a dramatic progress raises the question: how generalizable are neural networks in solving problems that demand broad skills? To answer this question, we propose SMART: a Simple Multimodal Algorithmic Reasoning Task and the associated SMART-101 dataset, for evaluating the abstraction, deduction, and generalization abilities of neural networks in solving visuo-linguistic puzzles designed specifically for children in the 6--8 age group. Our dataset consists of 101 unique puzzles; each puzzle comprises a picture and a question, and their solution needs a mix of several elementary skills, including arithmetic, algebra, and spatial reasoning, among others. To scale our dataset towards training deep neural networks, we programmatically generate entirely new instances for each puzzle, while retaining their solution algorithm. To benchmark performances on SMART-101, we propose a vision and language meta-learning model using varied state-of-the-art backbones. Our experiments reveal that while powerful deep models offer reasonable performances on puzzles in a supervised setting, they are not better than random accuracy when analyzed for generalization. We also evaluate the recent ChatGPT and other large language models on a part of SMART-101 and find that while these models show convincing reasoning abilities, the answers are often incorrect.
comment: Accepted at CVPR 2023. For the SMART-101 dataset, see https://doi.org/10.5281/zenodo.7761799
♻ ☆ Subject-driven Text-to-Image Generation via Apprenticeship Learning
Recent text-to-image generation models like DreamBooth have made remarkable progress in generating highly customized images of a target subject, by fine-tuning an ``expert model'' for a given subject from a few examples. However, this process is expensive, since a new expert model must be learned for each subject. In this paper, we present SuTI, a Subject-driven Text-to-Image generator that replaces subject-specific fine tuning with in-context learning. Given a few demonstrations of a new subject, SuTI can instantly generate novel renditions of the subject in different scenes, without any subject-specific optimization. SuTI is powered by apprenticeship learning, where a single apprentice model is learned from data generated by a massive number of subject-specific expert models. Specifically, we mine millions of image clusters from the Internet, each centered around a specific visual subject. We adopt these clusters to train a massive number of expert models, each specializing in a different subject. The apprentice model SuTI then learns to imitate the behavior of these fine-tuned experts. SuTI can generate high-quality and customized subject-specific images 20x faster than optimization-based SoTA methods. On the challenging DreamBench and DreamBench-v2, our human evaluation shows that SuTI significantly outperforms existing models like InstructPix2Pix, Textual Inversion, Imagic, Prompt2Prompt, Re-Imagen and DreamBooth, especially on the subject and text alignment aspects.
comment: Work in Progress
♻ ☆ Maximum Entropy on Erroneous Predictions (MEEP): Improving model calibration for medical image segmentation MICCAI 2023
Modern deep neural networks achieved remarkable progress in medical image segmentation tasks. However, it has recently been observed that they tend to produce overconfident estimates, even in situations of high uncertainty, leading to poorly calibrated and unreliable models. In this work we introduce Maximum Entropy on Erroneous Predictions (MEEP), a training strategy for segmentation networks which selectively penalizes overconfident predictions, focusing only on misclassified pixels. Our method is agnostic to the neural architecture, does not increase model complexity and can be coupled with multiple segmentation loss functions. We benchmark the proposed strategy in two challenging segmentation tasks: white matter hyperintensity lesions in magnetic resonance images (MRI) of the brain, and atrial segmentation in cardiac MRI. The experimental results demonstrate that coupling MEEP with standard segmentation losses leads to improvements not only in terms of model calibration, but also in segmentation quality.
comment: Accepted for publication at MICCAI 2023
♻ ☆ Can Deep Learning Reliably Recognize Abnormality Patterns on Chest X-rays? A Multi-Reader Study Examining One Month of AI Implementation in Everyday Radiology Clinical Practice
In this study, we developed a deep-learning-based automatic detection algorithm (DLAD, Carebot AI CXR) to detect and localize seven specific radiological findings (atelectasis (ATE), consolidation (CON), pleural effusion (EFF), pulmonary lesion (LES), subcutaneous emphysema (SCE), cardiomegaly (CMG), pneumothorax (PNO)) on chest X-rays (CXR). We collected 956 CXRs and compared the performance of the DLAD with that of six individual radiologists who assessed the images in a hospital setting. The proposed DLAD achieved high sensitivity (ATE 1.000 (0.624-1.000), CON 0.864 (0.671-0.956), EFF 0.953 (0.887-0.983), LES 0.905 (0.715-0.978), SCE 1.000 (0.366-1.000), CMG 0.837 (0.711-0.917), PNO 0.875 (0.538-0.986)), even when compared to the radiologists (LOWEST: ATE 0.000 (0.000-0.376), CON 0.182 (0.070-0.382), EFF 0.400 (0.302-0.506), LES 0.238 (0.103-0.448), SCE 0.000 (0.000-0.634), CMG 0.347 (0.228-0.486), PNO 0.375 (0.134-0.691), HIGHEST: ATE 1.000 (0.624-1.000), CON 0.864 (0.671-0.956), EFF 0.953 (0.887-0.983), LES 0.667 (0.456-0.830), SCE 1.000 (0.366-1.000), CMG 0.980 (0.896-0.999), PNO 0.875 (0.538-0.986)). The findings of the study demonstrate that the suggested DLAD holds potential for integration into everyday clinical practice as a decision support system, effectively mitigating the false negative rate associated with junior and intermediate radiologists.
♻ ☆ Matching-based Data Valuation for Generative Model
Data valuation is critical in machine learning, as it helps enhance model transparency and protect data properties. Existing data valuation methods have primarily focused on discriminative models, neglecting deep generative models that have recently gained considerable attention. Similar to discriminative models, there is an urgent need to assess data contributions in deep generative models as well. However, previous data valuation approaches mainly relied on discriminative model performance metrics and required model retraining. Consequently, they cannot be applied directly and efficiently to recent deep generative models, such as generative adversarial networks and diffusion models, in practice. To bridge this gap, we formulate the data valuation problem in generative models from a similarity-matching perspective. Specifically, we introduce Generative Model Valuator (GMValuator), the first model-agnostic approach for any generative models, designed to provide data valuation for generation tasks. We have conducted extensive experiments to demonstrate the effectiveness of the proposed method. To the best of their knowledge, GMValuator is the first work that offers a training-free, post-hoc data valuation strategy for deep generative models.
♻ ☆ Dynamic Spatial Sparsification for Efficient Vision Transformers and Convolutional Neural Networks NeurIPS 2021
In this paper, we present a new approach for model acceleration by exploiting spatial sparsity in visual data. We observe that the final prediction in vision Transformers is only based on a subset of the most informative tokens, which is sufficient for accurate image recognition. Based on this observation, we propose a dynamic token sparsification framework to prune redundant tokens progressively and dynamically based on the input to accelerate vision Transformers. Specifically, we devise a lightweight prediction module to estimate the importance score of each token given the current features. The module is added to different layers to prune redundant tokens hierarchically. While the framework is inspired by our observation of the sparse attention in vision Transformers, we find the idea of adaptive and asymmetric computation can be a general solution for accelerating various architectures. We extend our method to hierarchical models including CNNs and hierarchical vision Transformers as well as more complex dense prediction tasks that require structured feature maps by formulating a more generic dynamic spatial sparsification framework with progressive sparsification and asymmetric computation for different spatial locations. By applying lightweight fast paths to less informative features and using more expressive slow paths to more important locations, we can maintain the structure of feature maps while significantly reducing the overall computations. Extensive experiments demonstrate the effectiveness of our framework on various modern architectures and different visual recognition tasks. Our results clearly demonstrate that dynamic spatial sparsification offers a new and more effective dimension for model acceleration. Code is available at https://github.com/raoyongming/DynamicViT
comment: Accepted to T-PAMI. Journal version of our NeurIPS 2021 work: arXiv:2106.02034. Code is available at https://github.com/raoyongming/DynamicViT
♻ ☆ Fair Diffusion: Instructing Text-to-Image Generation Models on Fairness
Generative AI models have recently achieved astonishing results in quality and are consequently employed in a fast-growing number of applications. However, since they are highly data-driven, relying on billion-sized datasets randomly scraped from the internet, they also suffer from degenerated and biased human behavior, as we demonstrate. In fact, they may even reinforce such biases. To not only uncover but also combat these undesired effects, we present a novel strategy, called Fair Diffusion, to attenuate biases after the deployment of generative text-to-image models. Specifically, we demonstrate shifting a bias, based on human instructions, in any direction yielding arbitrarily new proportions for, e.g., identity groups. As our empirical evaluation demonstrates, this introduced control enables instructing generative image models on fairness, with no data filtering and additional training required.
♻ ☆ Motion-Based Sign Language Video Summarization using Curvature and Torsion
An interesting problem in many video-based applications is the generation of short synopses by selecting the most informative frames, a procedure which is known as video summarization. For sign language videos the benefits of using the $t$-parameterized counterpart of the curvature of the 2-D signer's wrist trajectory to identify keyframes, have been recently reported in the literature. In this paper we extend these ideas by modeling the 3-D hand motion that is extracted from each frame of the video. To this end we propose a new informative function based on the $t$-parameterized curvature and torsion of the 3-D trajectory. The method to characterize video frames as keyframes depends on whether the motion occurs in 2-D or 3-D space. Specifically, in the case of 3-D motion we look for the maxima of the harmonic mean of the curvature and torsion of the target's trajectory; in the planar motion case we seek for the maxima of the trajectory's curvature. The proposed 3-D feature is experimentally evaluated in applications of sign language videos on (1) objective measures using ground-truth keyframe annotations, (2) human-based evaluation of understanding, and (3) gloss classification and the results obtained are promising.
comment: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible
♻ ☆ Getting ViT in Shape: Scaling Laws for Compute-Optimal Model Design
Scaling laws have been recently employed to derive compute-optimal model size (number of parameters) for a given compute duration. We advance and refine such methods to infer compute-optimal model shapes, such as width and depth, and successfully implement this in vision transformers. Our shape-optimized vision transformer, SoViT, achieves results competitive with models that exceed twice its size, despite being pre-trained with an equivalent amount of compute. For example, SoViT-400m/14 achieves 90.3% fine-tuning accuracy on ILSRCV2012, surpassing the much larger ViT-g/14 and approaching ViT-G/14 under identical settings, with also less than half the inference cost. We conduct a thorough evaluation across multiple tasks, such as image classification, captioning, VQA and zero-shot transfer, demonstrating the effectiveness of our model across a broad range of domains and identifying limitations. Overall, our findings challenge the prevailing approach of blindly scaling up vision models and pave a path for a more informed scaling.
comment: 10 pages, 7 figures, 9 tables. Version 2: Layout fixes
♻ ☆ Architecture-Agnostic Masked Image Modeling -- From ViT back to CNN ICML 2023
Masked image modeling, an emerging self-supervised pre-training method, has shown impressive success across numerous downstream vision tasks with Vision transformers. Its underlying idea is simple: a portion of the input image is masked out and then reconstructed via a pre-text task. However, the working principle behind MIM is not well explained, and previous studies insist that MIM primarily works for the Transformer family but is incompatible with CNNs. In this work, we observe that MIM essentially teaches the model to learn better middle-order interactions among patches for more generalized feature extraction. We then propose an Architecture-Agnostic Masked Image Modeling framework (A$^2$MIM), which is compatible with both Transformers and CNNs in a unified way. Extensive experiments on popular benchmarks show that A$^2$MIM learns better representations without explicit design and endows the backbone model with the stronger capability to transfer to various downstream tasks.
comment: ICML 2023 (poster). The source code will be released in https://github.com/Westlake-AI/A2MIM
♻ ☆ HySpecNet-11k: A Large-Scale Hyperspectral Dataset for Benchmarking Learning-Based Hyperspectral Image Compression Methods
The development of learning-based hyperspectral image compression methods has recently attracted great attention in remote sensing. Such methods require a high number of hyperspectral images to be used during training to optimize all parameters and reach a high compression performance. However, existing hyperspectral datasets are not sufficient to train and evaluate learning-based compression methods, which hinders the research in this field. To address this problem, in this paper we present HySpecNet-11k that is a large-scale hyperspectral benchmark dataset made up of 11,483 nonoverlapping image patches. Each patch is a portion of 128 $\times$ 128 pixels with 224 spectral bands and a ground sample distance of 30 m. We exploit HySpecNet-11k to benchmark the current state of the art in learning-based hyperspectral image compression by focussing our attention on various 1D, 2D and 3D convolutional autoencoder architectures. Nevertheless, HySpecNet-11k can be used for any unsupervised learning task in the framework of hyperspectral image analysis. The dataset, our code and the pre-trained weights are publicly available at https://hyspecnet.rsim.berlin .
comment: Accepted at IEEE International Geoscience and Remote Sensing Symposium (IGARSS) 2023. The dataset, our code and the pre-trained weights are publicly available at https://hyspecnet.rsim.berlin
♻ ☆ Gradient Domain Weighted Guided Image Filtering
Guided image filter is a well-known local filter in image processing. However, the presence of halo artifacts is a common issue associated with this type of filter. This paper proposes an algorithm that utilizes gradient information to accurately identify the edges of an image. Furthermore, the algorithm uses weighted information to distinguish flat areas from edge areas, resulting in sharper edges and reduced blur in flat areas. This approach mitigates the excessive blurring near edges that often leads to halo artifacts. Experimental results demonstrate that the proposed algorithm significantly suppresses halo artifacts at the edges, making it highly effective for both image denoising and detail enhancement.
♻ ☆ LiT-4-RSVQA: Lightweight Transformer-based Visual Question Answering in Remote Sensing
Visual question answering (VQA) methods in remote sensing (RS) aim to answer natural language questions with respect to an RS image. Most of the existing methods require a large amount of computational resources, which limits their application in operational scenarios in RS. To address this issue, in this paper we present an effective lightweight transformer-based VQA in RS (LiT-4-RSVQA) architecture for efficient and accurate VQA in RS. Our architecture consists of: i) a lightweight text encoder module; ii) a lightweight image encoder module; iii) a fusion module; and iv) a classification module. The experimental results obtained on a VQA benchmark dataset demonstrate that our proposed LiT-4-RSVQA architecture provides accurate VQA results while significantly reducing the computational requirements on the executing hardware. Our code is publicly available at https://git.tu-berlin.de/rsim/lit4rsvqa.
comment: Accepted at IEEE International Geoscience and Remote Sensing Symposium 2023
♻ ☆ Your Contrastive Learning Is Secretly Doing Stochastic Neighbor Embedding ICLR 2023
Contrastive learning, especially self-supervised contrastive learning (SSCL), has achieved great success in extracting powerful features from unlabeled data. In this work, we contribute to the theoretical understanding of SSCL and uncover its connection to the classic data visualization method, stochastic neighbor embedding (SNE), whose goal is to preserve pairwise distances. From the perspective of preserving neighboring information, SSCL can be viewed as a special case of SNE with the input space pairwise similarities specified by data augmentation. The established correspondence facilitates deeper theoretical understanding of learned features of SSCL, as well as methodological guidelines for practical improvement. Specifically, through the lens of SNE, we provide novel analysis on domain-agnostic augmentations, implicit bias and robustness of learned features. To illustrate the practical advantage, we demonstrate that the modifications from SNE to $t$-SNE can also be adopted in the SSCL setting, achieving significant improvement in both in-distribution and out-of-distribution generalization.
comment: Accepted by ICLR 2023
♻ ☆ AvatarStudio: Text-driven Editing of 3D Dynamic Human Head Avatars
Capturing and editing full head performances enables the creation of virtual characters with various applications such as extended reality and media production. The past few years witnessed a steep rise in the photorealism of human head avatars. Such avatars can be controlled through different input data modalities, including RGB, audio, depth, IMUs and others. While these data modalities provide effective means of control, they mostly focus on editing the head movements such as the facial expressions, head pose and/or camera viewpoint. In this paper, we propose AvatarStudio, a text-based method for editing the appearance of a dynamic full head avatar. Our approach builds on existing work to capture dynamic performances of human heads using neural radiance field (NeRF) and edits this representation with a text-to-image diffusion model. Specifically, we introduce an optimization strategy for incorporating multiple keyframes representing different camera viewpoints and time stamps of a video performance into a single diffusion model. Using this personalized diffusion model, we edit the dynamic NeRF by introducing view-and-time-aware Score Distillation Sampling (VT-SDS) following a model-based guidance approach. Our method edits the full head in a canonical space, and then propagates these edits to remaining time steps via a pretrained deformation network. We evaluate our method visually and numerically via a user study, and results show that our method outperforms existing approaches. Our experiments validate the design choices of our method and highlight that our edits are genuine, personalized, as well as 3D- and time-consistent.
comment: 17 pages, 17 figures. Project page: https://vcai.mpi-inf.mpg.de/projects/AvatarStudio/
♻ ☆ SWL-Adapt: An Unsupervised Domain Adaptation Model with Sample Weight Learning for Cross-User Wearable Human Activity Recognition AAAI 2023
In practice, Wearable Human Activity Recognition (WHAR) models usually face performance degradation on the new user due to user variance. Unsupervised domain adaptation (UDA) becomes the natural solution to cross-user WHAR under annotation scarcity. Existing UDA models usually align samples across domains without differentiation, which ignores the difference among samples. In this paper, we propose an unsupervised domain adaptation model with sample weight learning (SWL-Adapt) for cross-user WHAR. SWL-Adapt calculates sample weights according to the classification loss and domain discrimination loss of each sample with a parameterized network. We introduce the meta-optimization based update rule to learn this network end-to-end, which is guided by meta-classification loss on the selected pseudo-labeled target samples. Therefore, this network can fit a weighting function according to the cross-user WHAR task at hand, which is superior to existing sample differentiation rules fixed for special scenarios. Extensive experiments on three public WHAR datasets demonstrate that SWL-Adapt achieves the state-of-the-art performance on the cross-user WHAR task, outperforming the best baseline by an average of 3.1% and 5.3% in accuracy and macro F1 score, respectively.
comment: Accepted by AAAI 2023. 9 pages and 4 figures in main text. 3 pages and 5 figures in appendix
♻ ☆ Primitive Simultaneous Optimization of Similarity Metrics for Image Registration
Even though simultaneous optimization of similarity metrics represents a standard procedure in the field of semantic segmentation, surprisingly, this does not hold true for image registration. To close this unexpected gap in the literature, we investigate in a complex multi-modal 3D setting whether simultaneous optimization of registration metrics, here implemented by means of primitive summation, can benefit image registration. We evaluate two challenging datasets containing collections of pre- to post-operative and pre- to intra-operative Magnetic Resonance Imaging (MRI) of glioma. Employing the proposed optimization we demonstrate improved registration accuracy in terms of Target Registration Error (TRE) on expert neuroradiologists' landmark annotations.
♻ ☆ CounTR: Transformer-based Generalised Visual Counting BMVC2022
In this paper, we consider the problem of generalised visual object counting, with the goal of developing a computational model for counting the number of objects from arbitrary semantic categories, using arbitrary number of "exemplars", i.e. zero-shot or few-shot counting. To this end, we make the following four contributions: (1) We introduce a novel transformer-based architecture for generalised visual object counting, termed as Counting Transformer (CounTR), which explicitly capture the similarity between image patches or with given "exemplars" with the attention mechanism;(2) We adopt a two-stage training regime, that first pre-trains the model with self-supervised learning, and followed by supervised fine-tuning;(3) We propose a simple, scalable pipeline for synthesizing training images with a large number of instances or that from different semantic categories, explicitly forcing the model to make use of the given "exemplars";(4) We conduct thorough ablation studies on the large-scale counting benchmark, e.g. FSC-147, and demonstrate state-of-the-art performance on both zero and few-shot settings.
comment: Accepted by BMVC2022
♻ ☆ Transformer-Based Visual Segmentation: A Survey
Visual segmentation seeks to partition images, video frames, or point clouds into multiple segments or groups. This technique has numerous real-world applications, such as autonomous driving, image editing, robot sensing, and medical analysis. Over the past decade, deep learning-based methods have made remarkable strides in this area. Recently, transformers, a type of neural network based on self-attention originally designed for natural language processing, have considerably surpassed previous convolutional or recurrent approaches in various vision processing tasks. Specifically, vision transformers offer robust, unified, and even simpler solutions for various segmentation tasks. This survey provides a thorough overview of transformer-based visual segmentation, summarizing recent advancements. We first review the background, encompassing problem definitions, datasets, and prior convolutional methods. Next, we summarize a meta-architecture that unifies all recent transformer-based approaches. Based on this meta-architecture, we examine various method designs, including modifications to the meta-architecture and associated applications. We also present several closely related settings, including 3D point cloud segmentation, foundation model tuning, domain-aware segmentation, efficient segmentation, and medical segmentation. Additionally, we compile and re-evaluate the reviewed methods on several well-established datasets. Finally, we identify open challenges in this field and propose directions for future research. The project page can be found at https://github.com/lxtGH/Awesome-Segmentation-With-Transformer. We will also continually monitor developments in this rapidly evolving field.
comment: Work in progress. Github: https://github.com/lxtGH/Awesome-Segmentation-With-Transformer
♻ ☆ RADIFUSION: A multi-radiomics deep learning based breast cancer risk prediction model using sequential mammographic images with image attention and bilateral asymmetry refinement
Breast cancer is a significant public health concern and early detection is critical for triaging high risk patients. Sequential screening mammograms can provide important spatiotemporal information about changes in breast tissue over time. In this study, we propose a deep learning architecture called RADIFUSION that utilizes sequential mammograms and incorporates a linear image attention mechanism, radiomic features, a new gating mechanism to combine different mammographic views, and bilateral asymmetry-based finetuning for breast cancer risk assessment. We evaluate our model on a screening dataset called Cohort of Screen-Aged Women (CSAW) dataset. Based on results obtained on the independent testing set consisting of 1,749 women, our approach achieved superior performance compared to other state-of-the-art models with area under the receiver operating characteristic curves (AUCs) of 0.905, 0.872 and 0.866 in the three respective metrics of 1-year AUC, 2-year AUC and > 2-year AUC. Our study highlights the importance of incorporating various deep learning mechanisms, such as image attention, radiomic features, gating mechanism, and bilateral asymmetry-based fine-tuning, to improve the accuracy of breast cancer risk assessment. We also demonstrate that our model's performance was enhanced by leveraging spatiotemporal information from sequential mammograms. Our findings suggest that RADIFUSION can provide clinicians with a powerful tool for breast cancer risk assessment.
comment: v2
♻ ☆ Multi-Granularity Archaeological Dating of Chinese Bronze Dings Based on a Knowledge-Guided Relation Graph CVPR2023
The archaeological dating of bronze dings has played a critical role in the study of ancient Chinese history. Current archaeology depends on trained experts to carry out bronze dating, which is time-consuming and labor-intensive. For such dating, in this study, we propose a learning-based approach to integrate advanced deep learning techniques and archaeological knowledge. To achieve this, we first collect a large-scale image dataset of bronze dings, which contains richer attribute information than other existing fine-grained datasets. Second, we introduce a multihead classifier and a knowledge-guided relation graph to mine the relationship between attributes and the ding era. Third, we conduct comparison experiments with various existing methods, the results of which show that our dating method achieves a state-of-the-art performance. We hope that our data and applied networks will enrich fine-grained classification research relevant to other interdisciplinary areas of expertise. The dataset and source code used are included in our supplementary materials, and will be open after submission owing to the anonymity policy. Source codes and data are available at: https://github.com/zhourixin/bronze-Ding.
comment: CVPR2023 accepted
♻ ☆ VGOS: Voxel Grid Optimization for View Synthesis from Sparse Inputs IJCAI 2023
Neural Radiance Fields (NeRF) has shown great success in novel view synthesis due to its state-of-the-art quality and flexibility. However, NeRF requires dense input views (tens to hundreds) and a long training time (hours to days) for a single scene to generate high-fidelity images. Although using the voxel grids to represent the radiance field can significantly accelerate the optimization process, we observe that for sparse inputs, the voxel grids are more prone to overfitting to the training views and will have holes and floaters, which leads to artifacts. In this paper, we propose VGOS, an approach for fast (3-5 minutes) radiance field reconstruction from sparse inputs (3-10 views) to address these issues. To improve the performance of voxel-based radiance field in sparse input scenarios, we propose two methods: (a) We introduce an incremental voxel training strategy, which prevents overfitting by suppressing the optimization of peripheral voxels in the early stage of reconstruction. (b) We use several regularization techniques to smooth the voxels, which avoids degenerate solutions. Experiments demonstrate that VGOS achieves state-of-the-art performance for sparse inputs with super-fast convergence. Code will be available at https://github.com/SJoJoK/VGOS.
comment: IJCAI 2023 Accepted (Main Track)
♻ ☆ SSD-MonoDETR: Supervised Scale-aware Deformable Transformer for Monocular 3D Object Detection
Transformer-based methods have demonstrated superior performance for monocular 3D object detection recently, which aims at predicting 3D attributes from a single 2D image. Most existing transformer-based methods leverage both visual and depth representations to explore valuable query points on objects, and the quality of the learned query points has a great impact on detection accuracy. Unfortunately, existing unsupervised attention mechanisms in transformers are prone to generate low-quality query features due to inaccurate receptive fields, especially on hard objects. To tackle this problem, this paper proposes a novel Supervised Scale-aware Deformable Attention (SSDA) for monocular 3D object detection. Specifically, SSDA presets several masks with different scales and utilizes depth and visual features to adaptively learn a scale-aware filter for object query augmentation. Imposing the scale awareness, SSDA could well predict the accurate receptive field of an object query to support robust query feature generation. Aside from this, SSDA is assigned with a Weighted Scale Matching (WSM) loss to supervise scale prediction, which presents more confident results as compared to the unsupervised attention mechanisms. Extensive experiments on the KITTI benchmark demonstrate that SSDA significantly improves the detection accuracy, especially on moderate and hard objects, yielding state-of-the-art performance as compared to the existing approaches. Our code will be made publicly available at https://github.com/mikasa3lili/SSD-MonoDETR.
comment: Code will be made publicly available at https://github.com/mikasa3lili/SSD-MonoDETR
♻ ☆ Beyond Pretrained Features: Noisy Image Modeling Provides Adversarial Defense
Recent advancements in masked image modeling (MIM) have made it a prevailing framework for self-supervised visual representation learning. The MIM pretrained models, like most deep neural network methods, are still vulnerable to adversarial attacks, limiting their practical application, and this issue has received little research attention. In this paper, we investigate how this powerful self-supervised learning paradigm can provide adversarial robustness to downstream classifiers. During the exploration, we find that noisy image modeling (NIM), a simple variant of MIM that adopts denoising as the pre-text task, reconstructs noisy images surprisingly well despite severe corruption. Motivated by this observation, we propose an adversarial defense method by exploiting the pretrained decoder for denoising, referred to as De^3, through which NIM is able to enhance adversarial robustness beyond providing pretrained features. Furthermore, we incorporate a simple modification, sampling the noise scale hyperparameter from random distributions, and enable the defense to achieve a better and tunable trade-off between accuracy and robustness. Experimental results demonstrate that, in terms of adversarial robustness, NIM is superior compared to MIM thanks to its effective denoising capability. Moreover, the defense provided by NIM achieves performance on par with adversarial training while offering the extra tunability advantage. Source code and models will be made available.
♻ ☆ Oscillation-free Quantization for Low-bit Vision Transformers
Weight oscillation is an undesirable side effect of quantization-aware training, in which quantized weights frequently jump between two quantized levels, resulting in training instability and a sub-optimal final model. We discover that the learnable scaling factor, a widely-used $\textit{de facto}$ setting in quantization aggravates weight oscillation. In this study, we investigate the connection between the learnable scaling factor and quantized weight oscillation and use ViT as a case driver to illustrate the findings and remedies. In addition, we also found that the interdependence between quantized weights in $\textit{query}$ and $\textit{key}$ of a self-attention layer makes ViT vulnerable to oscillation. We, therefore, propose three techniques accordingly: statistical weight quantization ($\rm StatsQ$) to improve quantization robustness compared to the prevalent learnable-scale-based method; confidence-guided annealing ($\rm CGA$) that freezes the weights with $\textit{high confidence}$ and calms the oscillating weights; and $\textit{query}$-$\textit{key}$ reparameterization ($\rm QKR$) to resolve the query-key intertwined oscillation and mitigate the resulting gradient misestimation. Extensive experiments demonstrate that these proposed techniques successfully abate weight oscillation and consistently achieve substantial accuracy improvement on ImageNet. Specifically, our 2-bit DeiT-T/DeiT-S algorithms outperform the previous state-of-the-art by 9.8% and 7.7%, respectively. Code and models are available at: https://github.com/nbasyl/OFQ.
comment: Proceedings of the 40 th International Conference on Machine Learning, Honolulu, Hawaii, USA. PMLR 202, 2023
♻ ☆ BKinD-3D: Self-Supervised 3D Keypoint Discovery from Multi-View Videos CVPR 2023
Quantifying motion in 3D is important for studying the behavior of humans and other animals, but manual pose annotations are expensive and time-consuming to obtain. Self-supervised keypoint discovery is a promising strategy for estimating 3D poses without annotations. However, current keypoint discovery approaches commonly process single 2D views and do not operate in the 3D space. We propose a new method to perform self-supervised keypoint discovery in 3D from multi-view videos of behaving agents, without any keypoint or bounding box supervision in 2D or 3D. Our method, BKinD-3D, uses an encoder-decoder architecture with a 3D volumetric heatmap, trained to reconstruct spatiotemporal differences across multiple views, in addition to joint length constraints on a learned 3D skeleton of the subject. In this way, we discover keypoints without requiring manual supervision in videos of humans and rats, demonstrating the potential of 3D keypoint discovery for studying behavior.
comment: CVPR 2023. Project page: https://sites.google.com/view/b-kind/3d Code: https://github.com/neuroethology/BKinD-3D
♻ ☆ An XAI Approach to Deep Learning Models in the Detection of DCIS
The results showed that XAI could indeed be used as a proof of concept to begin discussions on the implementation of assistive AI systems within the clinical community.
comment: 12 pages, 5 figures
♻ ☆ CLIP-Layout: Style-Consistent Indoor Scene Synthesis with Semantic Furniture Embedding
Indoor scene synthesis involves automatically picking and placing furniture appropriately on a floor plan, so that the scene looks realistic and is functionally plausible. Such scenes can serve as homes for immersive 3D experiences, or be used to train embodied agents. Existing methods for this task rely on labeled categories of furniture, e.g. bed, chair or table, to generate contextually relevant combinations of furniture. Whether heuristic or learned, these methods ignore instance-level visual attributes of objects, and as a result may produce visually less coherent scenes. In this paper, we introduce an auto-regressive scene model which can output instance-level predictions, using general purpose image embedding based on CLIP. This allows us to learn visual correspondences such as matching color and style, and produce more functionally plausible and aesthetically pleasing scenes. Evaluated on the 3D-FRONT dataset, our model achieves SOTA results in scene synthesis and improves auto-completion metrics by over 50%. Moreover, our embedding-based approach enables zero-shot text-guided scene synthesis and editing, which easily generalizes to furniture not seen during training.
comment: Changed paper template and cleaned up tables
♻ ☆ Learning Weakly Supervised Audio-Visual Violence Detection in Hyperbolic Space
In recent years, the task of weakly supervised audio-visual violence detection has gained considerable attention. The goal of this task is to identify violent segments within multimodal data based on video-level labels. Despite advances in this field, traditional Euclidean neural networks, which have been used in prior research, encounter difficulties in capturing highly discriminative representations due to limitations of the feature space. To overcome this, we propose HyperVD, a novel framework that learns snippet embeddings in hyperbolic space to improve model discrimination. Our framework comprises a detour fusion module for multimodal fusion, effectively alleviating modality inconsistency between audio and visual signals. Additionally, we contribute two branches of fully hyperbolic graph convolutional networks that excavate feature similarities and temporal relationships among snippets in hyperbolic space. By learning snippet representations in this space, the framework effectively learns semantic discrepancies between violent and normal events. Extensive experiments on the XD-Violence benchmark demonstrate that our method outperforms state-of-the-art methods by a sizable margin.
comment: 8 pages, 5 figures, typos are fixed
♻ ☆ Taming Self-Supervised Learning for Presentation Attack Detection: De-Folding and De-Mixing
Biometric systems are vulnerable to Presentation Attacks (PA) performed using various Presentation Attack Instruments (PAIs). Even though there are numerous Presentation Attack Detection (PAD) techniques based on both deep learning and hand-crafted features, the generalization of PAD for unknown PAI is still a challenging problem. In this work, we empirically prove that the initialization of the PAD model is a crucial factor for the generalization, which is rarely discussed in the community. Based on such observation, we proposed a self-supervised learning-based method, denoted as DF-DM. Specifically, DF-DM is based on a global-local view coupled with De-Folding and De-Mixing to derive the task-specific representation for PAD. During De-Folding, the proposed technique will learn region-specific features to represent samples in a local pattern by explicitly minimizing generative loss. While De-Mixing drives detectors to obtain the instance-specific features with global information for more comprehensive representation by minimizing interpolation-based consistency. Extensive experimental results show that the proposed method can achieve significant improvements in terms of both face and fingerprint PAD in more complicated and hybrid datasets when compared with state-of-the-art methods. When training in CASIA-FASD and Idiap Replay-Attack, the proposed method can achieve an 18.60% Equal Error Rate (EER) in OULU-NPU and MSU-MFSD, exceeding baseline performance by 9.54%. The source code of the proposed technique is available at https://github.com/kongzhecn/dfdm.
comment: Accepted by IEEE Transactions on Neural Networks and Learning Systems (TNNLS)
♻ ☆ MetaCLUE: Towards Comprehensive Visual Metaphors Research CVPR 2023
Creativity is an indispensable part of human cognition and also an inherent part of how we make sense of the world. Metaphorical abstraction is fundamental in communicating creative ideas through nuanced relationships between abstract concepts such as feelings. While computer vision benchmarks and approaches predominantly focus on understanding and generating literal interpretations of images, metaphorical comprehension of images remains relatively unexplored. Towards this goal, we introduce MetaCLUE, a set of vision tasks on visual metaphor. We also collect high-quality and rich metaphor annotations (abstract objects, concepts, relationships along with their corresponding object boxes) as there do not exist any datasets that facilitate the evaluation of these tasks. We perform a comprehensive analysis of state-of-the-art models in vision and language based on our annotations, highlighting strengths and weaknesses of current approaches in visual metaphor Classification, Localization, Understanding (retrieval, question answering, captioning) and gEneration (text-to-image synthesis) tasks. We hope this work provides a concrete step towards developing AI systems with human-like creative capabilities.
comment: Accepted in CVPR 2023. Project page: https://metaclue.github.io/ , Video summary: https://youtu.be/V3TmeNETL-o
♻ ☆ Differentiated Relevances Embedding for Group-based Referring Expression Comprehension
The key of referring expression comprehension lies in capturing the cross-modal visual-linguistic relevance. Existing works typically model the cross-modal relevance in each image, where the anchor object/expression and their positive expression/object have the same attribute as the negative expression/object, but with different attribute values. These objects/expressions are exclusively utilized to learn the implicit representation of the attribute by a pair of different values, which however impedes the accuracies of the attribute representations, expression/object representations, and their cross-modal relevances since each anchor object/expression usually has multiple attributes while each attribute usually has multiple potential values. To this end, we investigate a novel REC problem named Group-based REC, where each object/expression is simultaneously employed to construct the multiple triplets among the semantically similar images. To tackle the explosion of the negatives and the differentiation of the anchor-negative relevance scores, we propose the multi-group self-paced relevance learning schema to adaptively assign within-group object-expression pairs with different priorities based on their cross-modal relevances. Since the average cross-modal relevance varies a lot across different groups, we further design an across-group relevance constraint to balance the bias of the group priority. Experiments on three standard REC benchmarks demonstrate the effectiveness and superiority of our method.
♻ ☆ UMD: Unsupervised Model Detection for X2X Backdoor Attacks ICML 2023
Backdoor (Trojan) attack is a common threat to deep neural networks, where samples from one or more source classes embedded with a backdoor trigger will be misclassified to adversarial target classes. Existing methods for detecting whether a classifier is backdoor attacked are mostly designed for attacks with a single adversarial target (e.g., all-to-one attack). To the best of our knowledge, without supervision, no existing methods can effectively address the more general X2X attack with an arbitrary number of source classes, each paired with an arbitrary target class. In this paper, we propose UMD, the first Unsupervised Model Detection method that effectively detects X2X backdoor attacks via a joint inference of the adversarial (source, target) class pairs. In particular, we first define a novel transferability statistic to measure and select a subset of putative backdoor class pairs based on a proposed clustering approach. Then, these selected class pairs are jointly assessed based on an aggregation of their reverse-engineered trigger size for detection inference, using a robust and unsupervised anomaly detector we proposed. We conduct comprehensive evaluations on CIFAR-10, GTSRB, and Imagenette dataset, and show that our unsupervised UMD outperforms SOTA detectors (even with supervision) by 17%, 4%, and 8%, respectively, in terms of the detection accuracy against diverse X2X attacks. We also show the strong detection performance of UMD against several strong adaptive attacks.
comment: ICML 2023
Information Retrieval 17
☆ Fresh Content Needs More Attention: Multi-funnel Fresh Content Recommendation KDD 2023
Recommendation system serves as a conduit connecting users to an incredibly large, diverse and ever growing collection of contents. In practice, missing information on fresh (and tail) contents needs to be filled in order for them to be exposed and discovered by their audience. We here share our success stories in building a dedicated fresh content recommendation stack on a large commercial platform. To nominate fresh contents, we built a multi-funnel nomination system that combines (i) a two-tower model with strong generalization power for coverage, and (ii) a sequence model with near real-time update on user feedback for relevance. The multi-funnel setup effectively balances between coverage and relevance. An in-depth study uncovers the relationship between user activity level and their proximity toward fresh contents, which further motivates a contextual multi-funnel setup. Nominated fresh candidates are then scored and ranked by systems considering prediction uncertainty to further bootstrap content with less exposure. We evaluate the benefits of the dedicated fresh content recommendation stack, and the multi-funnel nomination system in particular, through user corpus co-diverted live experiments. We conduct multiple rounds of live experiments on a commercial platform serving billion of users demonstrating efficacy of our proposed methods.
comment: Accepted by KDD 2023
Pretrained Language Model based Web Search Ranking: From Relevance to Satisfaction
Search engine plays a crucial role in satisfying users' diverse information needs. Recently, Pretrained Language Models (PLMs) based text ranking models have achieved huge success in web search. However, many state-of-the-art text ranking approaches only focus on core relevance while ignoring other dimensions that contribute to user satisfaction, e.g., document quality, recency, authority, etc. In this work, we focus on ranking user satisfaction rather than relevance in web search, and propose a PLM-based framework, namely SAT-Ranker, which comprehensively models different dimensions of user satisfaction in a unified manner. In particular, we leverage the capacities of PLMs on both textual and numerical inputs, and apply a multi-field input that modularizes each dimension of user satisfaction as an input field. Overall, SAT-Ranker is an effective, extensible, and data-centric framework that has huge potential for industrial applications. On rigorous offline and online experiments, SAT-Ranker obtains remarkable gains on various evaluation sets targeting different dimensions of user satisfaction. It is now fully deployed online to improve the usability of our search engine.
☆ Influence Maximization with Fairness at Scale (Extended Version)
In this paper, we revisit the problem of influence maximization with fairness, which aims to select k influential nodes to maximise the spread of information in a network, while ensuring that selected sensitive user attributes are fairly affected, i.e., are proportionally similar between the original network and the affected users. Recent studies on this problem focused only on extremely small networks, hence the challenge remains on how to achieve a scalable solution, applicable to networks with millions or billions of nodes. We propose an approach that is based on learning node representations for fair spread from diffusion cascades, instead of the social connectivity s.t. we can deal with very large graphs. We propose two data-driven approaches: (a) fairness-based participant sampling (FPS), and (b) fairness as context (FAC). Spread related user features, such as the probability of diffusing information to others, are derived from the historical information cascades, using a deep neural network. The extracted features are then used in selecting influencers that maximize the influence spread, while being also fair with respect to the chosen sensitive attributes. In FPS, fairness and cascade length information are considered independently in the decision-making process, while FAC considers these information facets jointly and considers correlations between them. The proposed algorithms are generic and represent the first policy-driven solutions that can be applied to arbitrary sets of sensitive attributes at scale. We evaluate the performance of our solutions on a real-world public dataset (Sina Weibo) and on a hybrid real-synthethic dataset (Digg), which exhibit all the facets that we exploit, namely diffusion network, diffusion traces, and user profiles. These experiments show that our methods outperform the state-the-art solutions in terms of spread, fairness, and scalability.
☆ Système de recommandations basé sur les contraintes pour les simulations de gestion de crise
In the context of the evacuation of populations, some citizens/volunteers may want and be able to participate in the evacuation of populations in difficulty by coming to lend a hand to emergency/evacuation vehicles with their own vehicles. One way of framing these impulses of solidarity would be to be able to list in real-time the citizens/volunteers available with their vehicles (land, sea, air, etc.), to be able to geolocate them according to the risk areas to be evacuated, and adding them to the evacuation/rescue vehicles. Because it is difficult to propose an effective real-time operational system on the field in a real crisis situation, in this work, we propose to add a module for recommending driver/vehicle pairs (with their specificities) to a system of crisis management simulation. To do that, we chose to model and develop an ontology-supported constraint-based recommender system for crisis management simulations.
comment: in French language
☆ Hierarchical Reinforcement Learning for Modeling User Novelty-Seeking Intent in Recommender Systems
Recommending novel content, which expands user horizons by introducing them to new interests, has been shown to improve users' long-term experience on recommendation platforms \cite{chen2021values}. Users however are not constantly looking to explore novel content. It is therefore crucial to understand their novelty-seeking intent and adjust the recommendation policy accordingly. Most existing literature models a user's propensity to choose novel content or to prefer a more diverse set of recommendations at individual interactions. Hierarchical structure, on the other hand, exists in a user's novelty-seeking intent, which is manifested as a static and intrinsic user preference for seeking novelty along with a dynamic session-based propensity. To this end, we propose a novel hierarchical reinforcement learning-based method to model the hierarchical user novelty-seeking intent, and to adapt the recommendation policy accordingly based on the extracted user novelty-seeking propensity. We further incorporate diversity and novelty-related measurement in the reward function of the hierarchical RL (HRL) agent to encourage user exploration \cite{chen2021values}. We demonstrate the benefits of explicitly modeling hierarchical user novelty-seeking intent in recommendations through extensive experiments on simulated and real-world datasets. In particular, we demonstrate that the effectiveness of our proposed hierarchical RL-based method lies in its ability to capture such hierarchically-structured intent. As a result, the proposed HRL model achieves superior performance on several public datasets, compared with state-of-art baselines.
Prompt Tuning Large Language Models on Personalized Aspect Extraction for Recommendations
Existing aspect extraction methods mostly rely on explicit or ground truth aspect information, or using data mining or machine learning approaches to extract aspects from implicit user feedback such as user reviews. It however remains under-explored how the extracted aspects can help generate more meaningful recommendations to the users. Meanwhile, existing research on aspect-based recommendations often relies on separate aspect extraction models or assumes the aspects are given, without accounting for the fact the optimal set of aspects could be dependent on the recommendation task at hand. In this work, we propose to combine aspect extraction together with aspect-based recommendations in an end-to-end manner, achieving the two goals together in a single framework. For the aspect extraction component, we leverage the recent advances in large language models and design a new prompt learning mechanism to generate aspects for the end recommendation task. For the aspect-based recommendation component, the extracted aspects are concatenated with the usual user and item features used by the recommendation model. The recommendation task mediates the learning of the user embeddings and item embeddings, which are used as soft prompts to generate aspects. Therefore, the extracted aspects are personalized and contextualized by the recommendation task. We showcase the effectiveness of our proposed method through extensive experiments on three industrial datasets, where our proposed framework significantly outperforms state-of-the-art baselines in both the personalized aspect extraction and aspect-based recommendation tasks. In particular, we demonstrate that it is necessary and beneficial to combine the learning of aspect extraction and aspect-based recommendation together. We also conduct extensive ablation studies to understand the contribution of each design component in our framework.
☆ An OPC UA-based industrial Big Data architecture
Industry 4.0 factories are complex and data-driven. Data is yielded from many sources, including sensors, PLCs, and other devices, but also from IT, like ERP or CRM systems. We ask how to collect and process this data in a way, such that it includes metadata and can be used for industrial analytics or to derive intelligent support systems. This paper describes a new, query model based approach, which uses a big data architecture to capture data from various sources using OPC UA as a foundation. It buffers and preprocesses the information for the purpose of harmonizing and providing a holistic state space of a factory, as well as mappings to the current state of a production site. That information can be made available to multiple processing sinks, decoupled from the data sources, which enables them to work with the information without interfering with devices of the production, disturbing the network devices they are working in, or influencing the production process negatively. Metadata and connected semantic information is kept throughout the process, allowing to feed algorithms with meaningful data, so that it can be accessed in its entirety to perform time series analysis, machine learning or similar evaluations as well as replaying the data from the buffer for repeatable simulations.
☆ DWT-CompCNN: Deep Image Classification Network for High Throughput JPEG 2000 Compressed Documents
For any digital application with document images such as retrieval, the classification of document images becomes an essential stage. Conventionally for the purpose, the full versions of the documents, that is the uncompressed document images make the input dataset, which poses a threat due to the big volume required to accommodate the full versions of the documents. Therefore, it would be novel, if the same classification task could be accomplished directly (with some partial decompression) with the compressed representation of documents in order to make the whole process computationally more efficient. In this research work, a novel deep learning model, DWT CompCNN is proposed for classification of documents that are compressed using High Throughput JPEG 2000 (HTJ2K) algorithm. The proposed DWT-CompCNN comprises of five convolutional layers with filter sizes of 16, 32, 64, 128, and 256 consecutively for each increasing layer to improve learning from the wavelet coefficients extracted from the compressed images. Experiments are performed on two benchmark datasets- Tobacco-3482 and RVL-CDIP, which demonstrate that the proposed model is time and space efficient, and also achieves a better classification accuracy in compressed domain.
comment: In Springer Journal - Pattern Analysis and Applications under Minor Revision
☆ Reducing Popularity Bias in Recommender Systems through AUC-Optimal Negative Sampling
Popularity bias is a persistent issue associated with recommendation systems, posing challenges to both fairness and efficiency. Existing literature widely acknowledges that reducing popularity bias often requires sacrificing recommendation accuracy. In this paper, we challenge this commonly held belief. Our analysis under general bias-variance decomposition framework shows that reducing bias can actually lead to improved model performance under certain conditions. To achieve this win-win situation, we propose to intervene in model training through negative sampling thereby modifying model predictions. Specifically, we provide an optimal negative sampling rule that maximizes partial AUC to preserve the accuracy of any given model, while correcting sample information and prior information to reduce popularity bias in a flexible and principled way. Our experimental results on real-world datasets demonstrate the superiority of our approach in improving recommendation performance and reducing popularity bias.
comment: 20 pages
☆ LyricSIM: A novel Dataset and Benchmark for Similarity Detection in Spanish Song LyricS
In this paper, we present a new dataset and benchmark tailored to the task of semantic similarity in song lyrics. Our dataset, originally consisting of 2775 pairs of Spanish songs, was annotated in a collective annotation experiment by 63 native annotators. After collecting and refining the data to ensure a high degree of consensus and data integrity, we obtained 676 high-quality annotated pairs that were used to evaluate the performance of various state-of-the-art monolingual and multilingual language models. Consequently, we established baseline results that we hope will be useful to the community in all future academic and industrial applications conducted in this context.
comment: Accepted to Congreso Internacional de la Sociedad Espa\~nola para el Procesamiento del Lenguaje Natural 2023 (SEPLN2023)
☆ JEPOO: Highly Accurate Joint Estimation of Pitch, Onset and Offset for Music Information Retrieval IJCAI 2023
Melody extraction is a core task in music information retrieval, and the estimation of pitch, onset and offset are key sub-tasks in melody extraction. Existing methods have limited accuracy, and work for only one type of data, either single-pitch or multipitch. In this paper, we propose a highly accurate method for joint estimation of pitch, onset and offset, named JEPOO. We address the challenges of joint learning optimization and handling both single-pitch and multi-pitch data through novel model design and a new optimization technique named Pareto modulated loss with loss weight regularization. This is the first method that can accurately handle both single-pitch and multi-pitch music data, and even a mix of them. A comprehensive experimental study on a wide range of real datasets shows that JEPOO outperforms state-ofthe-art methods by up to 10.6%, 8.3% and 10.3% for the prediction of Pitch, Onset and Offset, respectively, and JEPOO is robust for various types of data and instruments. The ablation study shows the effectiveness of each component of JEPOO.
comment: This paper has been accepted by IJCAI 2023; 11 pages, 6 figures
☆ DeepScribe: Localization and Classification of Elamite Cuneiform Signs Via Deep Learning
Twenty-five hundred years ago, the paperwork of the Achaemenid Empire was recorded on clay tablets. In 1933, archaeologists from the University of Chicago's Oriental Institute (OI) found tens of thousands of these tablets and fragments during the excavation of Persepolis. Many of these tablets have been painstakingly photographed and annotated by expert cuneiformists, and now provide a rich dataset consisting of over 5,000 annotated tablet images and 100,000 cuneiform sign bounding boxes. We leverage this dataset to develop DeepScribe, a modular computer vision pipeline capable of localizing cuneiform signs and providing suggestions for the identity of each sign. We investigate the difficulty of learning subtasks relevant to cuneiform tablet transcription on ground-truth data, finding that a RetinaNet object detector can achieve a localization mAP of 0.78 and a ResNet classifier can achieve a top-5 sign classification accuracy of 0.89. The end-to-end pipeline achieves a top-5 classification accuracy of 0.80. As part of the classification module, DeepScribe groups cuneiform signs into morphological clusters. We consider how this automatic clustering approach differs from the organization of standard, printed sign lists and what we may learn from it. These components, trained individually, are sufficient to produce a system that can analyze photos of cuneiform tablets from the Achaemenid period and provide useful transliteration suggestions to researchers. We evaluate the model's end-to-end performance on locating and classifying signs, providing a roadmap to a linguistically-aware transliteration system, then consider the model's potential utility when applied to other periods of cuneiform writing.
comment: Currently under review in the ACM JOCCH
☆ Self Contrastive Learning for Session-based Recommendation
Session-based recommendation, which aims to predict the next item of users' interest as per an existing sequence interaction of items, has attracted growing applications of Contrastive Learning (CL) with improved user and item representations. However, these contrastive objectives: (1) serve a similar role as the cross-entropy loss while ignoring the item representation space optimisation; and (2) commonly require complicated modelling, including complex positive/negative sample constructions and extra data augmentation. In this work, we introduce Self-Contrastive Learning (SCL), which simplifies the application of CL and enhances the performance of state-of-the-art CL-based recommendation techniques. Specifically, SCL is formulated as an objective function that directly promotes a uniform distribution among item representations and efficiently replaces all the existing contrastive objective components of state-of-the-art models. Unlike previous works, SCL eliminates the need for any positive/negative sample construction or data augmentation, leading to enhanced interpretability of the item representation space and facilitating its extensibility to existing recommender systems. Through experiments on three benchmark datasets, we demonstrate that SCL consistently improves the performance of state-of-the-art models with statistical significance. Notably, our experiments show that SCL improves the performance of two best-performing models by 8.2% and 9.5% in P@10 (Precision) and 9.9% and 11.2% in MRR@10 (Mean Reciprocal Rank) on average across different benchmarks. Additionally, our analysis elucidates the improvement in terms of alignment and uniformity of representations, as well as the effectiveness of SCL with a low computational cost.
comment: Technical Report
☆ How Ready are Pre-trained Abstractive Models and LLMs for Legal Case Judgement Summarization?
Automatic summarization of legal case judgements has traditionally been attempted by using extractive summarization methods. However, in recent years, abstractive summarization models are gaining popularity since they can generate more natural and coherent summaries. Legal domain-specific pre-trained abstractive summarization models are now available. Moreover, general-domain pre-trained Large Language Models (LLMs), such as ChatGPT, are known to generate high-quality text and have the capacity for text summarization. Hence it is natural to ask if these models are ready for off-the-shelf application to automatically generate abstractive summaries for case judgements. To explore this question, we apply several state-of-the-art domain-specific abstractive summarization models and general-domain LLMs on Indian court case judgements, and check the quality of the generated summaries. In addition to standard metrics for summary quality, we check for inconsistencies and hallucinations in the summaries. We see that abstractive summarization models generally achieve slightly higher scores than extractive models in terms of standard summary evaluation metrics such as ROUGE and BLEU. However, we often find inconsistent or hallucinated information in the generated abstractive summaries. Overall, our investigation indicates that the pre-trained abstractive summarization models and LLMs are not yet ready for fully automatic deployment for case judgement summarization; rather a human-in-the-loop approach including manual checks for inconsistencies is more suitable at present.
comment: Accepted at the 3rd Workshop on Artificial Intelligence and Intelligent Assistance for Legal Professionals in the Digital Workplace (LegalAIIA 2023), in conjunction with the ICAIL 2023 conference
☆ Fast Interactive Search with a Scale-Free Comparison Oracle
A comparison-based search algorithm lets a user find a target item $t$ in a database by answering queries of the form, ``Which of items $i$ and $j$ is closer to $t$?'' Instead of formulating an explicit query (such as one or several keywords), the user navigates towards the target via a sequence of such (typically noisy) queries. We propose a scale-free probabilistic oracle model called $\gamma$-CKL for such similarity triplets $(i,j;t)$, which generalizes the CKL triplet model proposed in the literature. The generalization affords independent control over the discriminating power of the oracle and the dimension of the feature space containing the items. We develop a search algorithm with provably exponential rate of convergence under the $\gamma$-CKL oracle, thanks to a backtracking strategy that deals with the unavoidable errors in updating the belief region around the target. We evaluate the performance of the algorithm both over the posited oracle and over several real-world triplet datasets. We also report on a comprehensive user study, where human subjects navigate a database of face portraits.
♻ ☆ PrefRec: Recommender Systems with Human Preferences for Reinforcing Long-term User Engagement
Current advances in recommender systems have been remarkably successful in optimizing immediate engagement. However, long-term user engagement, a more desirable performance metric, remains difficult to improve. Meanwhile, recent reinforcement learning (RL) algorithms have shown their effectiveness in a variety of long-term goal optimization tasks. For this reason, RL is widely considered as a promising framework for optimizing long-term user engagement in recommendation. Though promising, the application of RL heavily relies on well-designed rewards, but designing rewards related to long-term user engagement is quite difficult. To mitigate the problem, we propose a novel paradigm, recommender systems with human preferences (or Preference-based Recommender systems), which allows RL recommender systems to learn from preferences about users historical behaviors rather than explicitly defined rewards. Such preferences are easily accessible through techniques such as crowdsourcing, as they do not require any expert knowledge. With PrefRec, we can fully exploit the advantages of RL in optimizing long-term goals, while avoiding complex reward engineering. PrefRec uses the preferences to automatically train a reward function in an end-to-end manner. The reward function is then used to generate learning signals to train the recommendation policy. Furthermore, we design an effective optimization method for PrefRec, which uses an additional value function, expectile regression and reward model pre-training to improve the performance. We conduct experiments on a variety of long-term user engagement optimization tasks. The results show that PrefRec significantly outperforms previous state-of-the-art methods in all the tasks.
♻ ☆ Self-Supervised Learning for Recommender Systems: A Survey
In recent years, neural architecture-based recommender systems have achieved tremendous success, but they still fall short of expectation when dealing with highly sparse data. Self-supervised learning (SSL), as an emerging technique for learning from unlabeled data, has attracted considerable attention as a potential solution to this issue. This survey paper presents a systematic and timely review of research efforts on self-supervised recommendation (SSR). Specifically, we propose an exclusive definition of SSR, on top of which we develop a comprehensive taxonomy to divide existing SSR methods into four categories: contrastive, generative, predictive, and hybrid. For each category, we elucidate its concept and formulation, the involved methods, as well as its pros and cons. Furthermore, to facilitate empirical comparison, we release an open-source library SELFRec (https://github.com/Coder-Yu/SELFRec), which incorporates a wide range of SSR models and benchmark datasets. Through rigorous experiments using this library, we derive and report some significant findings regarding the selection of self-supervised signals for enhancing recommendation. Finally, we shed light on the limitations in the current research and outline the future research directions.
comment: 20 pages. Accepted by TKDE
Machine Learning 150
☆ DaTaSeg: Taming a Universal Multi-Dataset Multi-Task Segmentation Model
Observing the close relationship among panoptic, semantic and instance segmentation tasks, we propose to train a universal multi-dataset multi-task segmentation model: DaTaSeg.We use a shared representation (mask proposals with class predictions) for all tasks. To tackle task discrepancy, we adopt different merge operations and post-processing for different tasks. We also leverage weak-supervision, allowing our segmentation model to benefit from cheaper bounding box annotations. To share knowledge across datasets, we use text embeddings from the same semantic embedding space as classifiers and share all network parameters among datasets. We train DaTaSeg on ADE semantic, COCO panoptic, and Objects365 detection datasets. DaTaSeg improves performance on all datasets, especially small-scale datasets, achieving 54.0 mIoU on ADE semantic and 53.5 PQ on COCO panoptic. DaTaSeg also enables weakly-supervised knowledge transfer on ADE panoptic and Objects365 instance segmentation. Experiments show DaTaSeg scales with the number of training datasets and enables open-vocabulary segmentation through direct transfer. In addition, we annotate an Objects365 instance segmentation set of 1,000 images and will release it as a public benchmark.
☆ DocFormerv2: Local Features for Document Understanding
We propose DocFormerv2, a multi-modal transformer for Visual Document Understanding (VDU). The VDU domain entails understanding documents (beyond mere OCR predictions) e.g., extracting information from a form, VQA for documents and other tasks. VDU is challenging as it needs a model to make sense of multiple modalities (visual, language and spatial) to make a prediction. Our approach, termed DocFormerv2 is an encoder-decoder transformer which takes as input - vision, language and spatial features. DocFormerv2 is pre-trained with unsupervised tasks employed asymmetrically i.e., two novel document tasks on encoder and one on the auto-regressive decoder. The unsupervised tasks have been carefully designed to ensure that the pre-training encourages local-feature alignment between multiple modalities. DocFormerv2 when evaluated on nine datasets shows state-of-the-art performance over strong baselines e.g. TabFact (4.3%), InfoVQA (1.4%), FUNSD (1%). Furthermore, to show generalization capabilities, on three VQA tasks involving scene-text, Doc- Formerv2 outperforms previous comparably-sized models and even does better than much larger models (such as GIT2, PaLi and Flamingo) on some tasks. Extensive ablations show that due to its pre-training, DocFormerv2 understands multiple modalities better than prior-art in VDU.
☆ PAGAR: Imitation Learning with Protagonist Antagonist Guided Adversarial Reward
Imitation learning (IL) algorithms often rely on inverse reinforcement learning (IRL) to first learn a reward function from expert demonstrations. However, IRL can suffer from identifiability issues and there is no performance or efficiency guarantee when training with the learned reward function. In this paper, we propose Protagonist Antagonist Guided Adversarial Reward (PAGAR), a semi-supervised learning paradigm for designing rewards for policy training. PAGAR employs an iterative adversarially search for reward functions to maximize the performance gap between a protagonist policy and an antagonist policy. This allows the protagonist policy to perform well across a set of possible reward functions despite the identifiability issue. When integrated with IRL-based IL, PAGAR guarantees that the trained policy succeeds in the underlying task. Furthermore, we introduce a practical on-and-off policy approach to IL with PAGAR. This approach maximally utilizes samples from both the protagonist and antagonist policies for the optimization of policy and reward functions. Experimental results demonstrate that our algorithm achieves higher training efficiency compared to state-of-the-art IL/IRL baselines in standard settings, as well as zero-shot learning from demonstrations in transfer environments.
☆ Broadcasting in random recursive dags
A uniform $k$-{\sc dag} generalizes the uniform random recursive tree by picking $k$ parents uniformly at random from the existing nodes. It starts with $k$ ''roots''. Each of the $k$ roots is assigned a bit. These bits are propagated by a noisy channel. The parents' bits are flipped with probability $p$, and a majority vote is taken. When all nodes have received their bits, the $k$-{\sc dag} is shown without identifying the roots. The goal is to estimate the majority bit among the roots. We identify the threshold for $p$ as a function of $k$ below which the majority rule among all nodes yields an error $c+o(1)$ with $c<1/2$. Above the threshold the majority rule errs with probability $1/2+o(1)$.
☆ Streaming algorithms for evaluating noisy judges on unlabeled data -- binary classification
The evaluation of noisy binary classifiers on unlabeled data is treated as a streaming task: given a data sketch of the decisions by an ensemble, estimate the true prevalence of the labels as well as each classifier's accuracy on them. Two fully algebraic evaluators are constructed to do this. Both are based on the assumption that the classifiers make independent errors. The first is based on majority voting. The second, the main contribution of the paper, is guaranteed to be correct. But how do we know the classifiers are independent on any given test? This principal/agent monitoring paradox is ameliorated by exploiting the failures of the independent evaluator to return sensible estimates. A search for nearly error independent trios is empirically carried out on the \texttt{adult}, \texttt{mushroom}, and \texttt{two-norm} datasets by using the algebraic failure modes to reject evaluation ensembles as too correlated. The searches are refined by constructing a surface in evaluation space that contains the true value point. The algebra of arbitrarily correlated classifiers permits the selection of a polynomial subset free of any correlation variables. Candidate evaluation ensembles are rejected if their data sketches produce independent estimates too far from the constructed surface. The results produced by the surviving ensembles can sometimes be as good as 1\%. But handling even small amounts of correlation remains a challenge. A Taylor expansion of the estimates produced when independence is assumed but the classifiers are, in fact, slightly correlated helps clarify how the independent evaluator has algebraic `blind spots'.
comment: 23 pages, 5 figures
☆ Graph Sparsification for GCN Towards Optimal Crop Yield Predictions
In agronomics, predicting crop yield at a per field/county granularity is important for farmers to minimize uncertainty and plan seeding for the next crop cycle. While state-of-the-art prediction techniques employ graph convolutional nets (GCN) to predict future crop yields given relevant features and crop yields of previous years, a dense underlying graph kernel requires long training and execution time. In this paper, we propose a graph sparsification method based on the Fiedler number to remove edges from a complete graph kernel, in order to lower the complexity of GCN training/execution. Specifically, we first show that greedily removing an edge at a time that induces the minimal change in the second eigenvalue leads to a sparse graph with good GCN performance. We then propose a fast method to choose an edge for removal per iteration based on an eigenvalue perturbation theorem. Experiments show that our Fiedler-based method produces a sparse graph with good GCN performance compared to other graph sparsification schemes in crop yield prediction.
☆ OMNI: Open-endedness via Models of human Notions of Interestingness
Open-ended algorithms aim to learn new, interesting behaviors forever. That requires a vast environment search space, but there are thus infinitely many possible tasks. Even after filtering for tasks the current agent can learn (i.e., learning progress), countless learnable yet uninteresting tasks remain (e.g., minor variations of previously learned tasks). An Achilles Heel of open-endedness research is the inability to quantify (and thus prioritize) tasks that are not just learnable, but also $\textit{interesting}$ (e.g., worthwhile and novel). We propose solving this problem by $\textit{Open-endedness via Models of human Notions of Interestingness}$ (OMNI). The insight is that we can utilize large (language) models (LMs) as a model of interestingness (MoI), because they $\textit{already}$ internalize human concepts of interestingness from training on vast amounts of human-generated data, where humans naturally write about what they find interesting or boring. We show that LM-based MoIs improve open-ended learning by focusing on tasks that are both learnable $\textit{and interesting}$, outperforming baselines based on uniform task sampling or learning progress alone. This approach has the potential to dramatically advance the ability to intelligently select which tasks to focus on next (i.e., auto-curricula), and could be seen as AI selecting its own next task to learn, facilitating self-improving AI and AI-Generating Algorithms.
comment: 33 pages, 22 figures
☆ A Data-Driven Measure of Relative Uncertainty for Misclassification Detection
Misclassification detection is an important problem in machine learning, as it allows for the identification of instances where the model's predictions are unreliable. However, conventional uncertainty measures such as Shannon entropy do not provide an effective way to infer the real uncertainty associated with the model's predictions. In this paper, we introduce a novel data-driven measure of relative uncertainty to an observer for misclassification detection. By learning patterns in the distribution of soft-predictions, our uncertainty measure can identify misclassified samples based on the predicted class probabilities. Interestingly, according to the proposed measure, soft-predictions that correspond to misclassified instances can carry a large amount of uncertainty, even though they may have low Shannon entropy. We demonstrate empirical improvements over multiple image classification tasks, outperforming state-of-the-art misclassification detection methods.
☆ Resolving Interference When Merging Models
Transfer learning - i.e., further fine-tuning a pre-trained model on a downstream task - can confer significant advantages, including improved downstream performance, faster convergence, and better sample efficiency. These advantages have led to a proliferation of task-specific fine-tuned models, which typically can only perform a single task and do not benefit from one another. Recently, model merging techniques have emerged as a solution to combine multiple task-specific models into a single multitask model without performing additional training. However, existing merging methods often ignore the interference between parameters of different models, resulting in large performance drops when merging multiple models. In this paper, we demonstrate that prior merging techniques inadvertently lose valuable information due to two major sources of interference: (a) interference due to redundant parameter values and (b) disagreement on the sign of a given parameter's values across models. To address this, we propose our method, TrIm, Elect Sign & Merge (TIES-Merging), which introduces three novel steps when merging models: (1) resetting parameters that only changed a small amount during fine-tuning, (2) resolving sign conflicts, and (3) merging only the parameters that are in alignment with the final agreed-upon sign. We find that TIES-Merging outperforms several existing methods in diverse settings covering a range of modalities, domains, number of tasks, model sizes, architectures, and fine-tuning settings. We further analyze the impact of different types of interference on model parameters, highlight the importance of resolving sign interference. Our code is available at https://github.com/prateeky2806/ties-merging
comment: 20 Pages, 10 Figures, 12 Tables
☆ Is Generative Modeling-based Stylization Necessary for Domain Adaptation in Regression Tasks?
Unsupervised domain adaptation (UDA) aims to bridge the gap between source and target domains in the absence of target domain labels using two main techniques: input-level alignment (such as generative modeling and stylization) and feature-level alignment (which matches the distribution of the feature maps, e.g. gradient reversal layers). Motivated from the success of generative modeling for image classification, stylization-based methods were recently proposed for regression tasks, such as pose estimation. However, use of input-level alignment via generative modeling and stylization incur additional overhead and computational complexity which limit their use in real-world DA tasks. To investigate the role of input-level alignment for DA, we ask the following question: Is generative modeling-based stylization necessary for visual domain adaptation in regression? Surprisingly, we find that input-alignment has little effect on regression tasks as compared to classification. Based on these insights, we develop a non-parametric feature-level domain alignment method -- Implicit Stylization (ImSty) -- which results in consistent improvements over SOTA regression task, without the need for computationally intensive stylization and generative modeling. Our work conducts a critical evaluation of the role of generative modeling and stylization, at a time when these are also gaining popularity for domain generalization.
☆ The Information Pathways Hypothesis: Transformers are Dynamic Self-Ensembles KDD23
Transformers use the dense self-attention mechanism which gives a lot of flexibility for long-range connectivity. Over multiple layers of a deep transformer, the number of possible connectivity patterns increases exponentially. However, very few of these contribute to the performance of the network, and even fewer are essential. We hypothesize that there are sparsely connected sub-networks within a transformer, called information pathways which can be trained independently. However, the dynamic (i.e., input-dependent) nature of these pathways makes it difficult to prune dense self-attention during training. But the overall distribution of these pathways is often predictable. We take advantage of this fact to propose Stochastically Subsampled self-Attention (SSA) - a general-purpose training strategy for transformers that can reduce both the memory and computational cost of self-attention by 4 to 8 times during training while also serving as a regularization method - improving generalization over dense training. We show that an ensemble of sub-models can be formed from the subsampled pathways within a network, which can achieve better performance than its densely attended counterpart. We perform experiments on a variety of NLP, computer vision and graph learning tasks in both generative and discriminative settings to provide empirical evidence for our claims and show the effectiveness of the proposed method.
comment: KDD23 preprint, 12 pages, 7 figures, 10 tables
☆ Affinity Clustering Framework for Data Debiasing Using Pairwise Distribution Discrepancy
Group imbalance, resulting from inadequate or unrepresentative data collection methods, is a primary cause of representation bias in datasets. Representation bias can exist with respect to different groups of one or more protected attributes and might lead to prejudicial and discriminatory outcomes toward certain groups of individuals; in cases where a learning model is trained on such biased data. This paper presents MASC, a data augmentation approach that leverages affinity clustering to balance the representation of non-protected and protected groups of a target dataset by utilizing instances of the same protected attributes from similar datasets that are categorized in the same cluster as the target dataset by sharing instances of the protected attribute. The proposed method involves constructing an affinity matrix by quantifying distribution discrepancies between dataset pairs and transforming them into a symmetric pairwise similarity matrix. A non-parametric spectral clustering is then applied to this affinity matrix, automatically categorizing the datasets into an optimal number of clusters. We perform a step-by-step experiment as a demo of our method to show the procedure of the proposed data augmentation method and evaluate and discuss its performance. A comparison with other data augmentation methods, both pre- and post-augmentation, is conducted, along with a model evaluation analysis of each method. Our method can handle non-binary protected attributes so, in our experiments, bias is measured in a non-binary protected attribute setup w.r.t. racial groups distribution for two separate minority groups in comparison with the majority group before and after debiasing. Empirical results imply that our method of augmenting dataset biases using real (genuine) data from similar contexts can effectively debias the target datasets comparably to existing data augmentation strategies.
comment: 15 pages plus 2 pages of references, 3 figures, 2 tables, and 1 algorithm
☆ MutateNN: Mutation Testing of Image Recognition Models Deployed on Hardware Accelerators
With the research advancement of Artificial Intelligence in the last years, there are new opportunities to mitigate real-world problems and advance technologically. Image recognition models in particular, are assigned with perception tasks to mitigate complex real-world challenges and lead to new solutions. Furthermore, the computational complexity and demand for resources of such models has also increased. To mitigate this, model optimization and hardware acceleration has come into play, but effectively integrating such concepts is a challenging and error-prone process. In order to allow developers and researchers to explore the robustness of deep learning image recognition models deployed on different hardware acceleration devices, we propose MutateNN, a tool that provides mutation testing and analysis capabilities for that purpose. To showcase its capabilities, we utilized 21 mutations for 7 widely-known pre-trained deep neural network models. We deployed our mutants on 4 different devices of varying computational capabilities and observed discrepancies in mutants related to conditional operations, as well as some unstable behaviour with those related to arithmetic types.
comment: 6 pages, 8 figures
☆ Evaluating Language Models for Mathematics through Interactions
The standard methodology of evaluating large language models (LLMs) based on static pairs of inputs and outputs is insufficient for developing assistants: this kind of assessments fails to take into account the essential interactive element in their deployment, and therefore limits how we understand language model capabilities. We introduce CheckMate, an adaptable prototype platform for humans to interact with and evaluate LLMs. We conduct a study with CheckMate to evaluate three language models~(InstructGPT, ChatGPT, and GPT-4) as assistants in proving undergraduate-level mathematics, with a mixed cohort of participants from undergraduate students to professors of mathematics. We release the resulting interaction and rating dataset, MathConverse. By analysing MathConverse, we derive a preliminary taxonomy of human behaviours and uncover that despite a generally positive correlation, there are notable instances of divergence between correctness and perceived helpfulness in LLM generations, amongst other findings. Further, we identify useful scenarios and existing issues of GPT-4 in mathematical reasoning through a series of case studies contributed by expert mathematicians. We conclude with actionable takeaways for ML practitioners and mathematicians: models which communicate uncertainty, respond well to user corrections, are more interpretable and concise may constitute better assistants; interactive evaluation is a promising way to continually navigate the capability of these models; humans should be aware of language models' algebraic fallibility, and for that reason discern where they should be used.
☆ Uniform Convergence of Deep Neural Networks with Lipschitz Continuous Activation Functions and Variable Widths
We consider deep neural networks with a Lipschitz continuous activation function and with weight matrices of variable widths. We establish a uniform convergence analysis framework in which sufficient conditions on weight matrices and bias vectors together with the Lipschitz constant are provided to ensure uniform convergence of the deep neural networks to a meaningful function as the number of their layers tends to infinity. In the framework, special results on uniform convergence of deep neural networks with a fixed width, bounded widths and unbounded widths are presented. In particular, as convolutional neural networks are special deep neural networks with weight matrices of increasing widths, we put forward conditions on the mask sequence which lead to uniform convergence of resulting convolutional neural networks. The Lipschitz continuity assumption on the activation functions allows us to include in our theory most of commonly used activation functions in applications.
☆ GateON: an unsupervised method for large scale continual learning
The objective of continual learning (CL) is to learn tasks sequentially without retraining on earlier tasks. However, when subjected to CL, traditional neural networks exhibit catastrophic forgetting and limited generalization. To overcome these problems, we introduce a novel method called 'Gate and Obstruct Network' (GateON). GateON combines learnable gating of activity and online estimation of parameter relevance to safeguard crucial knowledge from being overwritten. Our method generates partially overlapping pathways between tasks which permits forward and backward transfer during sequential learning. GateON addresses the issue of network saturation after parameter fixation by a re-activation mechanism of fixed neurons, enabling large-scale continual learning. GateON is implemented on a wide range of networks (fully-connected, CNN, Transformers), has low computational complexity, effectively learns up to 100 MNIST learning tasks, and achieves top-tier results for pre-trained BERT in CL-based NLP tasks.
☆ MKOR: Momentum-Enabled Kronecker-Factor-Based Optimizer Using Rank-1 Updates
This work proposes a Momentum-Enabled Kronecker-Factor-Based Optimizer Using Rank-1 updates, called MKOR, that improves the training time and convergence properties of deep neural networks (DNNs). Second-order techniques, while enjoying higher convergence rates vs first-order counterparts, have cubic complexity with respect to either the model size and/or the training batch size. Hence they exhibit poor scalability and performance in transformer models, e.g. large language models (LLMs), because the batch sizes in these models scale by the attention mechanism sequence length, leading to large model size and batch sizes. MKOR's complexity is quadratic with respect to the model size, alleviating the computation bottlenecks in second-order methods. Because of their high computation complexity, state-of-the-art implementations of second-order methods can only afford to update the second order information infrequently, and thus do not fully exploit the promise of better convergence from these updates. By reducing the communication complexity of the second-order updates as well as achieving a linear communication complexity, MKOR increases the frequency of second order updates. We also propose a hybrid version of MKOR (called MKOR-H) that mid-training falls backs to a first order optimizer if the second order updates no longer accelerate convergence. Our experiments show that MKOR outperforms state -of-the-art first order methods, e.g. the LAMB optimizer, and best implementations of second-order methods, i.e. KAISA/KFAC, up to 2.57x and 1.85x respectively on BERT-Large-Uncased on 64 GPUs.
☆ Harnessing large-language models to generate private synthetic text
Differentially private (DP) training methods like DP-SGD can protect sensitive training data by ensuring that ML models will not reveal private information. An alternative approach, which this paper studies, is to use a sensitive dataset to generate a new synthetic dataset which is differentially private with respect to the original data. Doing so has several advantages: synthetic data can be reused for other tasks (including for hyper parameter tuning), retained indefinitely, or shared with third parties without sacrificing privacy. However, obtaining DP data is much harder than introducing DP during training. To make it feasible for text, recent work has utilized public data by starting with a pre-trained generative language model and privately finetuning it on sensitive data. This model can be used to sample a DP synthetic dataset. While this strategy seems straightforward, executing it has proven problematic. Previous approaches either show significant performance loss, or have, as we show, critical design flaws. In this paper we demonstrate that a proper training objective along with tuning fewer parameters results in excellent DP synthetic data quality. Our approach is competitive with direct DP-training of downstream classifiers in terms of performance on downstream tasks. We also demonstrate that our DP synthetic data is not only useful for downstream classifier training, but also to tune those same models.
comment: 25 pages, 5 figures
☆ Balancing Exploration and Exploitation: Disentangled $β$-CVAE in De Novo Drug Design
Deep generative models have recently emerged as a promising de novo drug design method. In this respect, deep generative conditional variational autoencoder (CVAE) models are a powerful approach for generating novel molecules with desired drug-like properties. However, molecular graph-based models with disentanglement and multivariate explicit latent conditioning have not been fully elucidated. To address this, we proposed a molecular-graph $\beta$-CVAE model for de novo drug design. Here, we empirically tuned the value of disentanglement and assessed its ability to generate molecules with optimised univariate- or-multivariate properties. In particular, we optimised the octanol-water partition coefficient (ClogP), molar refractivity (CMR), quantitative estimate of drug-likeness (QED), and synthetic accessibility score (SAS). Results suggest that a lower $\beta$ value increases the uniqueness of generated molecules (exploration). Univariate optimisation results showed our model generated molecular property averages of ClogP = 41.07% $\pm$ 0.01% and CMR 66.76% $\pm$ 0.01% by the Ghose filter. Multivariate property optimisation results showed that our model generated an average of 30.07% $\pm$ 0.01% molecules for both desired properties. Furthermore, our model improved the QED and SAS (exploitation) of molecules generated. Together, these results suggest that the $\beta$-CVAE could balance exploration and exploitation through disentanglement and is a promising model for de novo drug design, thus providing a basis for future studies.
☆ Neural Differential Recurrent Neural Network with Adaptive Time Steps
The neural Ordinary Differential Equation (ODE) model has shown success in learning complex continuous-time processes from observations on discrete time stamps. In this work, we consider the modeling and forecasting of time series data that are non-stationary and may have sharp changes like spikes. We propose an RNN-based model, called RNN-ODE-Adap, that uses a neural ODE to represent the time development of the hidden states, and we adaptively select time steps based on the steepness of changes of the data over time so as to train the model more efficiently for the "spike-like" time series. Theoretically, RNN-ODE-Adap yields provably a consistent estimation of the intensity function for the Hawkes-type time series data. We also provide an approximation analysis of the RNN-ODE model showing the benefit of adaptive steps. The proposed model is demonstrated to achieve higher prediction accuracy with reduced computational cost on simulated dynamic system data and point process data and on a real electrocardiography dataset.
☆ Enhancing CLIP with CLIP: Exploring Pseudolabeling for Limited-Label Prompt Tuning
Fine-tuning vision-language models (VLMs) like CLIP to downstream tasks is often necessary to optimize their performance. However, a major obstacle is the limited availability of labeled data. We study the use of pseudolabels, i.e., heuristic labels for unlabeled data, to enhance CLIP via prompt tuning. Conventional pseudolabeling trains a model on labeled data and then generates labels for unlabeled data. VLMs' zero-shot capabilities enable a ``second generation'' of pseudolabeling approaches that do not require task-specific training on labeled data. By using zero-shot pseudolabels as a source of supervision, we observe that learning paradigms such as semi-supervised, transductive zero-shot, and unsupervised learning can all be seen as optimizing the same loss function. This unified view enables the development of versatile training strategies that are applicable across learning paradigms. We investigate them on image classification tasks where CLIP exhibits limitations, by varying prompt modalities, e.g., textual or visual prompts, and learning paradigms. We find that (1) unexplored prompt tuning strategies that iteratively refine pseudolabels consistently improve CLIP accuracy, by 19.5 points in semi-supervised learning, by 28.4 points in transductive zero-shot learning, and by 15.2 points in unsupervised learning, and (2) unlike conventional semi-supervised pseudolabeling, which exacerbates model biases toward classes with higher-quality pseudolabels, prompt tuning leads to a more equitable distribution of per-class accuracy. The code to reproduce the experiments is at github.com/BatsResearch/menghini-enhanceCLIPwithCLIP-code.
☆ XAI Renaissance: Redefining Interpretability in Medical Diagnostic Models
As machine learning models become increasingly prevalent in medical diagnostics, the need for interpretability and transparency becomes paramount. The XAI Renaissance signifies a significant shift in the field, aiming to redefine the interpretability of medical diagnostic models. This paper explores the innovative approaches and methodologies within the realm of Explainable AI (XAI) that are revolutionizing the interpretability of medical diagnostic models. By shedding light on the underlying decision-making process, XAI techniques empower healthcare professionals to understand, trust, and effectively utilize these models for accurate and reliable medical diagnoses. This review highlights the key advancements in XAI for medical diagnostics and their potential to transform the healthcare landscape, ultimately improving patient outcomes and fostering trust in AI-driven diagnostic systems.
☆ An Adaptive Method for Weak Supervision with Drifting Data
We introduce an adaptive method with formal quality guarantees for weak supervision in a non-stationary setting. Our goal is to infer the unknown labels of a sequence of data by using weak supervision sources that provide independent noisy signals of the correct classification for each data point. This setting includes crowdsourcing and programmatic weak supervision. We focus on the non-stationary case, where the accuracy of the weak supervision sources can drift over time, e.g., because of changes in the underlying data distribution. Due to the drift, older data could provide misleading information to infer the label of the current data point. Previous work relied on a priori assumptions on the magnitude of the drift to decide how much data to use from the past. Comparatively, our algorithm does not require any assumptions on the drift, and it adapts based on the input. In particular, at each step, our algorithm guarantees an estimation of the current accuracies of the weak supervision sources over a window of past observations that minimizes a trade-off between the error due to the variance of the estimation and the error due to the drift. Experiments on synthetic and real-world labelers show that our approach indeed adapts to the drift. Unlike fixed-window-size strategies, it dynamically chooses a window size that allows it to consistently maintain good performance.
☆ Poisoning Network Flow Classifiers
As machine learning (ML) classifiers increasingly oversee the automated monitoring of network traffic, studying their resilience against adversarial attacks becomes critical. This paper focuses on poisoning attacks, specifically backdoor attacks, against network traffic flow classifiers. We investigate the challenging scenario of clean-label poisoning where the adversary's capabilities are constrained to tampering only with the training data - without the ability to arbitrarily modify the training labels or any other component of the training process. We describe a trigger crafting strategy that leverages model interpretability techniques to generate trigger patterns that are effective even at very low poisoning rates. Finally, we design novel strategies to generate stealthy triggers, including an approach based on generative Bayesian network models, with the goal of minimizing the conspicuousness of the trigger, and thus making detection of an ongoing poisoning campaign more challenging. Our findings provide significant insights into the feasibility of poisoning attacks on network traffic classifiers used in multiple scenarios, including detecting malicious communication and application classification.
comment: 14 pages, 8 figures
☆ GANs Settle Scores!
Generative adversarial networks (GANs) comprise a generator, trained to learn the underlying distribution of the desired data, and a discriminator, trained to distinguish real samples from those output by the generator. A majority of GAN literature focuses on understanding the optimality of the discriminator through integral probability metric (IPM) or divergence based analysis. In this paper, we propose a unified approach to analyzing the generator optimization through variational approach. In $f$-divergence-minimizing GANs, we show that the optimal generator is the one that matches the score of its output distribution with that of the data distribution, while in IPM GANs, we show that this optimal generator matches score-like functions, involving the flow-field of the kernel associated with a chosen IPM constraint space. Further, the IPM-GAN optimization can be seen as one of smoothed score-matching, where the scores of the data and the generator distributions are convolved with the kernel associated with the constraint. The proposed approach serves to unify score-based training and existing GAN flavors, leveraging results from normalizing flows, while also providing explanations for empirical phenomena such as the stability of non-saturating GAN losses. Based on these results, we propose novel alternatives to $f$-GAN and IPM-GAN training based on score and flow matching, and discriminator-guided Langevin sampling.
☆ Fair multilingual vandalism detection system for Wikipedia
This paper presents a novel design of the system aimed at supporting the Wikipedia community in addressing vandalism on the platform. To achieve this, we collected a massive dataset of 47 languages, and applied advanced filtering and feature engineering techniques, including multilingual masked language modeling to build the training dataset from human-generated data. The performance of the system was evaluated through comparison with the one used in production in Wikipedia, known as ORES. Our research results in a significant increase in the number of languages covered, making Wikipedia patrolling more efficient to a wider range of communities. Furthermore, our model outperforms ORES, ensuring that the results provided are not only more accurate but also less biased against certain groups of contributors.
☆ Federated Multi-Sequence Stochastic Approximation with Local Hypergradient Estimation
Stochastic approximation with multiple coupled sequences (MSA) has found broad applications in machine learning as it encompasses a rich class of problems including bilevel optimization (BLO), multi-level compositional optimization (MCO), and reinforcement learning (specifically, actor-critic methods). However, designing provably-efficient federated algorithms for MSA has been an elusive question even for the special case of double sequence approximation (DSA). Towards this goal, we develop FedMSA which is the first federated algorithm for MSA, and establish its near-optimal communication complexity. As core novelties, (i) FedMSA enables the provable estimation of hypergradients in BLO and MCO via local client updates, which has been a notable bottleneck in prior theory, and (ii) our convergence guarantees are sensitive to the heterogeneity-level of the problem. We also incorporate momentum and variance reduction techniques to achieve further acceleration leading to near-optimal rates. Finally, we provide experiments that support our theory and demonstrate the empirical benefits of FedMSA. As an example, FedMSA enables order-of-magnitude savings in communication rounds compared to prior federated BLO schemes.
☆ Auditing for Human Expertise
High-stakes prediction tasks (e.g., patient diagnosis) are often handled by trained human experts. A common source of concern about automation in these settings is that experts may exercise intuition that is difficult to model and/or have access to information (e.g., conversations with a patient) that is simply unavailable to a would-be algorithm. This raises a natural question whether human experts add value which could not be captured by an algorithmic predictor. We develop a statistical framework under which we can pose this question as a natural hypothesis test. Indeed, as our framework highlights, detecting human expertise is more subtle than simply comparing the accuracy of expert predictions to those made by a particular learning algorithm. Instead, we propose a simple procedure which tests whether expert predictions are statistically independent from the outcomes of interest after conditioning on the available inputs (`features'). A rejection of our test thus suggests that human experts may add value to any algorithm trained on the available data, and has direct implications for whether human-AI `complementarity' is achievable in a given prediction task. We highlight the utility of our procedure using admissions data collected from the emergency department of a large academic hospital system, where we show that physicians' admit/discharge decisions for patients with acute gastrointestinal bleeding (AGIB) appear to be incorporating information not captured in a standard algorithmic screening tool. This is despite the fact that the screening tool is arguably more accurate than physicians' discretionary decisions, highlighting that -- even absent normative concerns about accountability or interpretability -- accuracy is insufficient to justify algorithmic automation.
comment: 27 pages, 8 figures
☆ Reduction of finite sampling noise in quantum neural networks
Quantum neural networks (QNNs) use parameterized quantum circuits with data-dependent inputs and generate outputs through the evaluation of expectation values. Calculating these expectation values necessitates repeated circuit evaluations, thus introducing fundamental finite-sampling noise even on error-free quantum computers. We reduce this noise by introducing the variance regularization, a technique for reducing the variance of the expectation value during the quantum model training. This technique requires no additional circuit evaluations if the QNN is properly constructed. Our empirical findings demonstrate the reduced variance speeds up the training and lowers the output noise as well as decreases the number of measurements in the gradient circuit evaluation. This regularization method is benchmarked on the regression of multiple functions. We show that in our examples, it lowers the variance by an order of magnitude on average and leads to a significantly reduced noise level of the QNN. We finally demonstrate QNN training on a real quantum device and evaluate the impact of error mitigation. Here, the optimization is practical only due to the reduced number shots in the gradient evaluation resulting from the reduced variance.
comment: 11 pages, 10 figures
☆ Do we become wiser with time? On causal equivalence with tiered background knowledge UAI 2023
Equivalence classes of DAGs (represented by CPDAGs) may be too large to provide useful causal information. Here, we address incorporating tiered background knowledge yielding restricted equivalence classes represented by 'tiered MPDAGs'. Tiered knowledge leads to considerable gains in informativeness and computational efficiency: We show that construction of tiered MPDAGs only requires application of Meek's 1st rule, and that tiered MPDAGs (unlike general MPDAGs) are chain graphs with chordal components. This entails simplifications e.g. of determining valid adjustment sets for causal effect estimation. Further, we characterise when one tiered ordering is more informative than another, providing insights into useful aspects of background knowledge.
comment: Accepted for the 39th Conference on Uncertainty in Artificial Intelligence (UAI 2023)
☆ Gode -- Integrating Biochemical Knowledge Graph into Pre-training Molecule Graph Neural Network
The precise prediction of molecular properties holds paramount importance in facilitating the development of innovative treatments and comprehending the intricate interplay between chemicals and biological systems. In this study, we propose a novel approach that integrates graph representations of individual molecular structures with multi-domain information from biomedical knowledge graphs (KGs). Integrating information from both levels, we can pre-train a more extensive and robust representation for both molecule-level and KG-level prediction tasks with our novel self-supervision strategy. For performance evaluation, we fine-tune our pre-trained model on 11 challenging chemical property prediction tasks. Results from our framework demonstrate our fine-tuned models outperform existing state-of-the-art models.
comment: It's an ongoing work. We're exploring the ability of Gode on other tasks
☆ HomE: Homography-Equivariant Video Representation Learning
Recent advances in self-supervised representation learning have enabled more efficient and robust model performance without relying on extensive labeled data. However, most works are still focused on images, with few working on videos and even fewer on multi-view videos, where more powerful inductive biases can be leveraged for self-supervision. In this work, we propose a novel method for representation learning of multi-view videos, where we explicitly model the representation space to maintain Homography Equivariance (HomE). Our method learns an implicit mapping between different views, culminating in a representation space that maintains the homography relationship between neighboring views. We evaluate our HomE representation via action recognition and pedestrian intent prediction as downstream tasks. On action classification, our method obtains 96.4% 3-fold accuracy on the UCF101 dataset, better than most state-of-the-art self-supervised learning methods. Similarly, on the STIP dataset, we outperform the state-of-the-art by 6% for pedestrian intent prediction one second into the future while also obtaining an accuracy of 91.2% for pedestrian action (cross vs. not-cross) classification. Code is available at https://github.com/anirudhs123/HomE.
comment: 10 pages, 4 figures, 4 tables
☆ Analyzing Credit Risk Model Problems through NLP-Based Clustering and Machine Learning: Insights from Validation Reports
This paper explores the use of clustering methods and machine learning algorithms, including Natural Language Processing (NLP), to identify and classify problems identified in credit risk models through textual information contained in validation reports. Using a unique dataset of 657 findings raised by validation teams in a large international banking group between January 2019 and December 2022. The findings are classified into nine validation dimensions and assigned a severity level by validators using their expert knowledge. The authors use embedding generation for the findings' titles and observations using four different pre-trained models, including "module\_url" from TensorFlow Hub and three models from the SentenceTransformer library, namely "all-mpnet-base-v2", "all-MiniLM-L6-v2", and "paraphrase-mpnet-base-v2". The paper uses and compares various clustering methods in grouping findings with similar characteristics, enabling the identification of common problems within each validation dimension and severity. The results of the study show that clustering is an effective approach for identifying and classifying credit risk model problems with accuracy higher than 60\%. The authors also employ machine learning algorithms, including logistic regression and XGBoost, to predict the validation dimension and its severity, achieving an accuracy of 80\% for XGBoost algorithm. Furthermore, the study identifies the top 10 words that predict a validation dimension and severity. Overall, this paper makes a contribution by demonstrating the usefulness of clustering and machine learning for analyzing textual information in validation reports, and providing insights into the types of problems encountered in the development and validation of credit risk models.
comment: 8 pages, 7 tables, 2 figures
☆ Hyperparameter Learning under Data Poisoning: Analysis of the Influence of Regularization via Multiobjective Bilevel Optimization
Machine Learning (ML) algorithms are vulnerable to poisoning attacks, where a fraction of the training data is manipulated to deliberately degrade the algorithms' performance. Optimal attacks can be formulated as bilevel optimization problems and help to assess their robustness in worst-case scenarios. We show that current approaches, which typically assume that hyperparameters remain constant, lead to an overly pessimistic view of the algorithms' robustness and of the impact of regularization. We propose a novel optimal attack formulation that considers the effect of the attack on the hyperparameters and models the attack as a multiobjective bilevel optimization problem. This allows to formulate optimal attacks, learn hyperparameters and evaluate robustness under worst-case conditions. We apply this attack formulation to several ML classifiers using $L_2$ and $L_1$ regularization. Our evaluation on multiple datasets confirms the limitations of previous strategies and evidences the benefits of using $L_2$ and $L_1$ regularization to dampen the effect of poisoning attacks.
comment: Accepted to IEEE Transactions on Neural Networks and Learning Systems. arXiv admin note: text overlap with arXiv:2003.00040
☆ Centered Self-Attention Layers
The self-attention mechanism in transformers and the message-passing mechanism in graph neural networks are repeatedly applied within deep learning architectures. We show that this application inevitably leads to oversmoothing, i.e., to similar representations at the deeper layers for different tokens in transformers and different nodes in graph neural networks. Based on our analysis, we present a correction term to the aggregating operator of these mechanisms. Empirically, this simple term eliminates much of the oversmoothing problem in visual transformers, obtaining performance in weakly supervised segmentation that surpasses elaborate baseline methods that introduce multiple auxiliary networks and training phrases. In graph neural networks, the correction term enables the training of very deep architectures more effectively than many recent solutions to the same problem.
☆ Decentralized Federated Learning: A Survey and Perspective
Federated learning (FL) has been gaining attention for its ability to share knowledge while maintaining user data, protecting privacy, increasing learning efficiency, and reducing communication overhead. Decentralized FL (DFL) is a decentralized network architecture that eliminates the need for a central server in contrast to centralized FL (CFL). DFL enables direct communication between clients, resulting in significant savings in communication resources. In this paper, a comprehensive survey and profound perspective is provided for DFL. First, a review of the methodology, challenges, and variants of CFL is conducted, laying the background of DFL. Then, a systematic and detailed perspective on DFL is introduced, including iteration order, communication protocols, network topologies, paradigm proposals, and temporal variability. Next, based on the definition of DFL, several extended variants and categorizations are proposed with state-of-the-art technologies. Lastly, in addition to summarizing the current challenges in the DFL, some possible solutions and future research directions are also discussed.
☆ Transfer learning for atomistic simulations using GNNs and kernel mean embeddings
Interatomic potentials learned using machine learning methods have been successfully applied to atomistic simulations. However, deep learning pipelines are notoriously data-hungry, while generating reference calculations is computationally demanding. To overcome this difficulty, we propose a transfer learning algorithm that leverages the ability of graph neural networks (GNNs) in describing chemical environments, together with kernel mean embeddings. We extract a feature map from GNNs pre-trained on the OC20 dataset and use it to learn the potential energy surface from system-specific datasets of catalytic processes. Our method is further enhanced by a flexible kernel function that incorporates chemical species information, resulting in improved performance and interpretability. We test our approach on a series of realistic datasets of increasing complexity, showing excellent generalization and transferability performance, and improving on methods that rely on GNNs or ridge regression alone, as well as similar fine-tuning approaches. We make the code available to the community at https://github.com/IsakFalk/atomistic_transfer_mekrr.
comment: 18 pages, 3 figures, 5 tables
☆ Probabilistic Concept Bottleneck Models ICML
Interpretable models are designed to make decisions in a human-interpretable manner. Representatively, Concept Bottleneck Models (CBM) follow a two-step process of concept prediction and class prediction based on the predicted concepts. CBM provides explanations with high-level concepts derived from concept predictions; thus, reliable concept predictions are important for trustworthiness. In this study, we address the ambiguity issue that can harm reliability. While the existence of a concept can often be ambiguous in the data, CBM predicts concepts deterministically without considering this ambiguity. To provide a reliable interpretation against this ambiguity, we propose Probabilistic Concept Bottleneck Models (ProbCBM). By leveraging probabilistic concept embeddings, ProbCBM models uncertainty in concept prediction and provides explanations based on the concept and its corresponding uncertainty. This uncertainty enhances the reliability of the explanations. Furthermore, as class uncertainty is derived from concept uncertainty in ProbCBM, we can explain class uncertainty by means of concept uncertainty. Code is publicly available at https://github.com/ejkim47/prob-cbm.
comment: International Conference on Machine Learning (ICML) 2023
☆ Spatio-Temporal Deep Learning-Assisted Reduced Security-Constrained Unit Commitment
Security-constrained unit commitment (SCUC) is a computationally complex process utilized in power system day-ahead scheduling and market clearing. SCUC is run daily and requires state-of-the-art algorithms to speed up the process. The constraints and data associated with SCUC are both geographically and temporally correlated to ensure the reliability of the solution, which further increases the complexity. In this paper, an advanced machine learning (ML) model is used to study the patterns in power system historical data, which inherently considers both spatial and temporal (ST) correlations in constraints. The ST-correlated ML model is trained to understand spatial correlation by considering graph neural networks (GNN) whereas temporal sequences are studied using long short-term memory (LSTM) networks. The proposed approach is validated on several test systems namely, IEEE 24-Bus system, IEEE-73 Bus system, IEEE 118-Bus system, and synthetic South-Carolina (SC) 500-Bus system. Moreover, B-{\theta} and power transfer distribution factor (PTDF) based SCUC formulations were considered in this research. Simulation results demonstrate that the ST approach can effectively predict generator commitment schedule and classify critical and non-critical lines in the system which are utilized for model reduction of SCUC to obtain computational enhancement without loss in solution quality
comment: 8 Figures, 5 Tables, 1 Algorithm
☆ Publicly available datasets of breast histopathology H&E whole-slide images: A systematic review
Advancements in digital pathology and computing resources have made a significant impact in the field of computational pathology for breast cancer diagnosis and treatment. However, access to high-quality labeled histopathological images of breast cancer is a big challenge that limits the development of accurate and robust deep learning models. In this systematic review, we identified the publicly available datasets of breast H&E stained whole-slide images (WSI) that can be used to develop deep learning algorithms. We systematically searched nine scientific literature databases and nine research data repositories. We found twelve publicly available datasets, containing 5153 H&E WSIs of breast cancer. Moreover, we reported image metadata and characteristics for each dataset to assist researchers in selecting proper datasets for specific tasks in breast cancer computational pathology. In addition, we compiled a list of patch and private datasets that were used in the included articles as a supplementary resource for researchers. Notably, 22% of the included articles utilized multiple datasets, and only 12% of the articles used an external validation set, suggesting that the performance of other developed models may be susceptible to overestimation. The TCGA-BRCA was used in 47.4% of the selected studies. This dataset has a considerable selection bias that can impact the robustness and generalizability of the trained algorithms. There is also a lack of consistent metadata reporting of breast WSI datasets that can be an issue in developing accurate deep learning models, indicating the necessity of establishing explicit guidelines for documenting breast WSI dataset characteristics and metadata.
comment: 27 pages (including references), 6 figures, 3 tables, 4 supporting information materials
☆ Does it pay to optimize AUC? AAAI
The Area Under the ROC Curve (AUC) is an important model metric for evaluating binary classifiers, and many algorithms have been proposed to optimize AUC approximately. It raises the question of whether the generally insignificant gains observed by previous studies are due to inherent limitations of the metric or the inadequate quality of optimization. To better understand the value of optimizing for AUC, we present an efficient algorithm, namely AUC-opt, to find the provably optimal AUC linear classifier in $\mathbb{R}^2$, which runs in $\mathcal{O}(n_+ n_- \log (n_+ n_-))$ where $n_+$ and $n_-$ are the number of positive and negative samples respectively. Furthermore, it can be naturally extended to $\mathbb{R}^d$ in $\mathcal{O}((n_+n_-)^{d-1}\log (n_+n_-))$ by calling AUC-opt in lower-dimensional spaces recursively. We prove the problem is NP-complete when $d$ is not fixed, reducing from the \textit{open hemisphere problem}. Experiments show that compared with other methods, AUC-opt achieves statistically significant improvements on between 17 to 40 in $\mathbb{R}^2$ and between 4 to 42 in $\mathbb{R}^3$ of 50 t-SNE training datasets. However, generally the gain proves insignificant on most testing datasets compared to the best standard classifiers. Similar observations are found for nonlinear AUC methods under real-world datasets.
comment: 16 pages, AAAI
Transformer-based Multi-Modal Learning for Multi Label Remote Sensing Image Classification
In this paper, we introduce a novel Synchronized Class Token Fusion (SCT Fusion) architecture in the framework of multi-modal multi-label classification (MLC) of remote sensing (RS) images. The proposed architecture leverages modality-specific attention-based transformer encoders to process varying input modalities, while exchanging information across modalities by synchronizing the special class tokens after each transformer encoder block. The synchronization involves fusing the class tokens with a trainable fusion transformation, resulting in a synchronized class token that contains information from all modalities. As the fusion transformation is trainable, it allows to reach an accurate representation of the shared features among different modalities. Experimental results show the effectiveness of the proposed architecture over single-modality architectures and an early fusion multi-modal architecture when evaluated on a multi-modal MLC dataset. The code of the proposed architecture is publicly available at https://git.tu-berlin.de/rsim/sct-fusion.
comment: Accepted at IEEE International Geoscience and Remote Sensing Symposium 2023
☆ Network Degeneracy as an Indicator of Training Performance: Comparing Finite and Infinite Width Angle Predictions
Neural networks are powerful functions with widespread use, but the theoretical behaviour of these functions is not fully understood. Creating deep neural networks by stacking many layers has achieved exceptional performance in many applications and contributed to the recent explosion of these methods. Previous works have shown that depth can exponentially increase the expressibility of the network. However, as networks get deeper and deeper, they are more susceptible to becoming degenerate. We observe this degeneracy in the sense that on initialization, inputs tend to become more and more correlated as they travel through the layers of the network. If a network has too many layers, it tends to approximate a (random) constant function, making it effectively incapable of distinguishing between inputs. This seems to affect the training of the network and cause it to perform poorly, as we empirically investigate in this paper. We use a simple algorithm that can accurately predict the level of degeneracy for any given fully connected ReLU network architecture, and demonstrate how the predicted degeneracy relates to training dynamics of the network. We also compare this prediction to predictions derived using infinite width networks.
comment: 5 pages, comments welcome
☆ Supervised Adversarial Contrastive Learning for Emotion Recognition in Conversations ACL 2023
Extracting generalized and robust representations is a major challenge in emotion recognition in conversations (ERC). To address this, we propose a supervised adversarial contrastive learning (SACL) framework for learning class-spread structured representations. The framework applies contrast-aware adversarial training to generate worst-case samples and uses a joint class-spread contrastive learning objective on both original and adversarial samples. It can effectively utilize label-level feature consistency and retain fine-grained intra-class features. To avoid the negative impact of adversarial perturbations on context-dependent data, we design a contextual adversarial training strategy to learn more diverse features from context and enhance the model's context robustness. We develop a sequence-based method SACL-LSTM under this framework, to learn label-consistent and context-robust emotional features for ERC. Experiments on three datasets demonstrate that SACL-LSTM achieves state-of-the-art performance on ERC. Extended experiments prove the effectiveness of the SACL framework.
comment: 16 pages, accepted by ACL 2023
☆ Can LLMs like GPT-4 outperform traditional AI tools in dementia diagnosis? Maybe, but not today
Recent investigations show that large language models (LLMs), specifically GPT-4, not only have remarkable capabilities in common Natural Language Processing (NLP) tasks but also exhibit human-level performance on various professional and academic benchmarks. However, whether GPT-4 can be directly used in practical applications and replace traditional artificial intelligence (AI) tools in specialized domains requires further experimental validation. In this paper, we explore the potential of LLMs such as GPT-4 to outperform traditional AI tools in dementia diagnosis. Comprehensive comparisons between GPT-4 and traditional AI tools are conducted to examine their diagnostic accuracy in a clinical setting. Experimental results on two real clinical datasets show that, although LLMs like GPT-4 demonstrate potential for future advancements in dementia diagnosis, they currently do not surpass the performance of traditional AI tools. The interpretability and faithfulness of GPT-4 are also evaluated by comparison with real doctors. We discuss the limitations of GPT-4 in its current state and propose future research directions to enhance GPT-4 in dementia diagnosis.
comment: 16 pages, 6 figures
☆ Local Message Passing on Frustrated Systems UAI 2023
Message passing on factor graphs is a powerful framework for probabilistic inference, which finds important applications in various scientific domains. The most wide-spread message passing scheme is the sum-product algorithm (SPA) which gives exact results on trees but often fails on graphs with many small cycles. We search for an alternative message passing algorithm that works particularly well on such cyclic graphs. Therefore, we challenge the extrinsic principle of the SPA, which loses its objective on graphs with cycles. We further replace the local SPA message update rule at the factor nodes of the underlying graph with a generic mapping, which is optimized in a data-driven fashion. These modifications lead to a considerable improvement in performance while preserving the simplicity of the SPA. We evaluate our method for two classes of cyclic graphs: the 2x2 fully connected Ising grid and factor graphs for symbol detection on linear communication channels with inter-symbol interference. To enable the method for large graphs as they occur in practical applications, we develop a novel loss function that is inspired by the Bethe approximation from statistical physics and allows for training in an unsupervised fashion.
comment: To appear at UAI 2023
☆ On Feature Diversity in Energy-based Models
Energy-based learning is a powerful learning paradigm that encapsulates various discriminative and generative approaches. An energy-based model (EBM) is typically formed of inner-model(s) that learn a combination of the different features to generate an energy mapping for each input configuration. In this paper, we focus on the diversity of the produced feature set. We extend the probably approximately correct (PAC) theory of EBMs and analyze the effect of redundancy reduction on the performance of EBMs. We derive generalization bounds for various learning contexts, i.e., regression, classification, and implicit regression, with different energy functions and we show that indeed reducing redundancy of the feature set can consistently decrease the gap between the true and empirical expectation of the energy and boosts the performance of the model.
comment: 18 pages, 3 figures
☆ Robust low-rank training via approximate orthonormal constraints
With the growth of model and data sizes, a broad effort has been made to design pruning techniques that reduce the resource demand of deep learning pipelines, while retaining model performance. In order to reduce both inference and training costs, a prominent line of work uses low-rank matrix factorizations to represent the network weights. Although able to retain accuracy, we observe that low-rank methods tend to compromise model robustness against adversarial perturbations. By modeling robustness in terms of the condition number of the neural network, we argue that this loss of robustness is due to the exploding singular values of the low-rank weight matrices. Thus, we introduce a robust low-rank training algorithm that maintains the network's weights on the low-rank matrix manifold while simultaneously enforcing approximate orthonormal constraints. The resulting model reduces both training and inference costs while ensuring well-conditioning and thus better adversarial robustness, without compromising model accuracy. This is shown by extensive numerical evidence and by our main approximation theorem that shows the computed robust low-rank network well-approximates the ideal full model, provided a highly performing low-rank sub-network exists.
☆ Hierarchical Reinforcement Learning for Modeling User Novelty-Seeking Intent in Recommender Systems
Recommending novel content, which expands user horizons by introducing them to new interests, has been shown to improve users' long-term experience on recommendation platforms \cite{chen2021values}. Users however are not constantly looking to explore novel content. It is therefore crucial to understand their novelty-seeking intent and adjust the recommendation policy accordingly. Most existing literature models a user's propensity to choose novel content or to prefer a more diverse set of recommendations at individual interactions. Hierarchical structure, on the other hand, exists in a user's novelty-seeking intent, which is manifested as a static and intrinsic user preference for seeking novelty along with a dynamic session-based propensity. To this end, we propose a novel hierarchical reinforcement learning-based method to model the hierarchical user novelty-seeking intent, and to adapt the recommendation policy accordingly based on the extracted user novelty-seeking propensity. We further incorporate diversity and novelty-related measurement in the reward function of the hierarchical RL (HRL) agent to encourage user exploration \cite{chen2021values}. We demonstrate the benefits of explicitly modeling hierarchical user novelty-seeking intent in recommendations through extensive experiments on simulated and real-world datasets. In particular, we demonstrate that the effectiveness of our proposed hierarchical RL-based method lies in its ability to capture such hierarchically-structured intent. As a result, the proposed HRL model achieves superior performance on several public datasets, compared with state-of-art baselines.
Prompt Tuning Large Language Models on Personalized Aspect Extraction for Recommendations
Existing aspect extraction methods mostly rely on explicit or ground truth aspect information, or using data mining or machine learning approaches to extract aspects from implicit user feedback such as user reviews. It however remains under-explored how the extracted aspects can help generate more meaningful recommendations to the users. Meanwhile, existing research on aspect-based recommendations often relies on separate aspect extraction models or assumes the aspects are given, without accounting for the fact the optimal set of aspects could be dependent on the recommendation task at hand. In this work, we propose to combine aspect extraction together with aspect-based recommendations in an end-to-end manner, achieving the two goals together in a single framework. For the aspect extraction component, we leverage the recent advances in large language models and design a new prompt learning mechanism to generate aspects for the end recommendation task. For the aspect-based recommendation component, the extracted aspects are concatenated with the usual user and item features used by the recommendation model. The recommendation task mediates the learning of the user embeddings and item embeddings, which are used as soft prompts to generate aspects. Therefore, the extracted aspects are personalized and contextualized by the recommendation task. We showcase the effectiveness of our proposed method through extensive experiments on three industrial datasets, where our proposed framework significantly outperforms state-of-the-art baselines in both the personalized aspect extraction and aspect-based recommendation tasks. In particular, we demonstrate that it is necessary and beneficial to combine the learning of aspect extraction and aspect-based recommendation together. We also conduct extensive ablation studies to understand the contribution of each design component in our framework.
☆ Generalist Equivariant Transformer Towards 3D Molecular Interaction Learning
Many processes in biology and drug discovery involve various 3D interactions between different molecules, such as protein and protein, protein and small molecule, etc. Designing a generalist model to learn universal molecular interactions is valuable yet challenging, given that different molecules are usually represented in different granularity. In this paper, we first propose to universally represent a 3D molecule as a geometric graph of sets, in contrast to conventional single-level representations. Upon the proposed unified representation, we then propose a Generalist Equivariant Transformer (GET) to effectively capture both sparse block-level and dense atom-level interactions. To be specific, GET consists of a bilevel attention module, a feed-forward module and a layer normalization module, where, notably, each module is E(3) equivariant to meet the symmetry of 3D world. Extensive experiments on the prediction of protein-protein affinity, ligand binding affinity, and ligand efficacy prediction verify the effectiveness of our proposed method against existing methods, and reveal its potential to learn transferable knowledge across different domains and different tasks.
comment: preprint
☆ Guiding Text-to-Text Privatization by Syntax
Metric Differential Privacy is a generalization of differential privacy tailored to address the unique challenges of text-to-text privatization. By adding noise to the representation of words in the geometric space of embeddings, words are replaced with words located in the proximity of the noisy representation. Since embeddings are trained based on word co-occurrences, this mechanism ensures that substitutions stem from a common semantic context. Without considering the grammatical category of words, however, this mechanism cannot guarantee that substitutions play similar syntactic roles. We analyze the capability of text-to-text privatization to preserve the grammatical category of words after substitution and find that surrogate texts consist almost exclusively of nouns. Lacking the capability to produce surrogate texts that correlate with the structure of the sensitive texts, we encompass our analysis by transforming the privatization step into a candidate selection problem in which substitutions are directed to words with matching grammatical properties. We demonstrate a substantial improvement in the performance of downstream tasks by up to $4.66\%$ while retaining comparative privacy guarantees.
☆ MLP-Mixer as a Wide and Sparse MLP
Multi-layer perceptron (MLP) is a fundamental component of deep learning that has been extensively employed for various problems. However, recent empirical successes in MLP-based architectures, particularly the progress of the MLP-Mixer, have revealed that there is still hidden potential in improving MLPs to achieve better performance. In this study, we reveal that the MLP-Mixer works effectively as a wide MLP with certain sparse weights. Initially, we clarify that the mixing layer of the Mixer has an effective expression as a wider MLP whose weights are sparse and represented by the Kronecker product. This expression naturally defines a permuted-Kronecker (PK) family, which can be regarded as a general class of mixing layers and is also regarded as an approximation of Monarch matrices. Subsequently, because the PK family effectively constitutes a wide MLP with sparse weights, one can apply the hypothesis proposed by Golubeva, Neyshabur and Gur-Ari (2021) that the prediction performance improves as the width (sparsity) increases when the number of weights is fixed. We empirically verify this hypothesis by maximizing the effective width of the MLP-Mixer, which enables us to determine the appropriate size of the mixing layers quantitatively.
comment: 19 pages, 13 figures
☆ GANs and alternative methods of synthetic noise generation for domain adaption of defect classification of Non-destructive ultrasonic testing
This work provides a solution to the challenge of small amounts of training data in Non-Destructive Ultrasonic Testing for composite components. It was demonstrated that direct simulation alone is ineffective at producing training data that was representative of the experimental domain due to poor noise reconstruction. Therefore, four unique synthetic data generation methods were proposed which use semi-analytical simulated data as a foundation. Each method was evaluated on its classification performance of real experimental images when trained on a Convolutional Neural Network which underwent hyperparameter optimization using a genetic algorithm. The first method introduced task specific modifications to CycleGAN, to learn the mapping from physics-based simulations of defect indications to experimental indications in resulting ultrasound images. The second method was based on combining real experimental defect free images with simulated defect responses. The final two methods fully simulated the noise responses at an image and signal level respectively. The purely simulated data produced a mean classification F1 score of 0.394. However, when trained on the new synthetic datasets, a significant improvement in classification performance on experimental data was realized, with mean classification F1 scores of 0.843, 0.688, 0.629, and 0.738 for the respective approaches.
comment: 16 Pages
☆ Theoretical Behavior of XAI Methods in the Presence of Suppressor Variables ICML 2023
In recent years, the community of 'explainable artificial intelligence' (XAI) has created a vast body of methods to bridge a perceived gap between model 'complexity' and 'interpretability'. However, a concrete problem to be solved by XAI methods has not yet been formally stated. As a result, XAI methods are lacking theoretical and empirical evidence for the 'correctness' of their explanations, limiting their potential use for quality-control and transparency purposes. At the same time, Haufe et al. (2014) showed, using simple toy examples, that even standard interpretations of linear models can be highly misleading. Specifically, high importance may be attributed to so-called suppressor variables lacking any statistical relation to the prediction target. This behavior has been confirmed empirically for a large array of XAI methods in Wilming et al. (2022). Here, we go one step further by deriving analytical expressions for the behavior of a variety of popular XAI methods on a simple two-dimensional binary classification problem involving Gaussian class-conditional distributions. We show that the majority of the studied approaches will attribute non-zero importance to a non-class-related suppressor feature in the presence of correlated noise. This poses important limitations on the interpretations and conclusions that the outputs of these XAI methods can afford.
comment: Accepted at ICML 2023
☆ ReLU to the Rescue: Improve Your On-Policy Actor-Critic with Positive Advantages
In this paper, we introduce a novel method for enhancing the effectiveness of on-policy Deep Reinforcement Learning (DRL) algorithms. Current on-policy algorithms, such as Proximal Policy Optimization (PPO) and Asynchronous Advantage Actor-Critic (A3C), do not sufficiently account for cautious interaction with the environment. Our method addresses this gap by explicitly integrating cautious interaction in two critical ways: by maximizing a lower-bound on the true value function plus a constant, thereby promoting a \textit{conservative value estimation}, and by incorporating Thompson sampling for cautious exploration. These features are realized through three surprisingly simple modifications to the A3C algorithm: processing advantage estimates through a ReLU function, spectral normalization, and dropout. We provide theoretical proof that our algorithm maximizes the lower bound, which also grounds Regret Matching Policy Gradients (RMPG), a discrete-action on-policy method for multi-agent reinforcement learning. Our rigorous empirical evaluations across various benchmarks consistently demonstrates our approach's improved performance against existing on-policy algorithms. This research represents a substantial step towards more cautious and effective DRL algorithms, which has the potential to unlock application to complex, real-world problems.
☆ Driving Context into Text-to-Text Privatization
\textit{Metric Differential Privacy} enables text-to-text privatization by adding calibrated noise to the vector of a word derived from an embedding space and projecting this noisy vector back to a discrete vocabulary using a nearest neighbor search. Since words are substituted without context, this mechanism is expected to fall short at finding substitutes for words with ambiguous meanings, such as \textit{'bank'}. To account for these ambiguous words, we leverage a sense embedding and incorporate a sense disambiguation step prior to noise injection. We encompass our modification to the privatization mechanism with an estimation of privacy and utility. For word sense disambiguation on the \textit{Words in Context} dataset, we demonstrate a substantial increase in classification accuracy by $6.05\%$.
☆ Unsupervised Paraphrasing of Multiword Expressions ACL 2023
We propose an unsupervised approach to paraphrasing multiword expressions (MWEs) in context. Our model employs only monolingual corpus data and pre-trained language models (without fine-tuning), and does not make use of any external resources such as dictionaries. We evaluate our method on the SemEval 2022 idiomatic semantic text similarity task, and show that it outperforms all unsupervised systems and rivals supervised systems.
comment: 13 pages; accepted for Findings of ACL 2023
☆ Towards Robust FastSpeech 2 by Modelling Residual Multimodality INTERSPEECH 2023
State-of-the-art non-autoregressive text-to-speech (TTS) models based on FastSpeech 2 can efficiently synthesise high-fidelity and natural speech. For expressive speech datasets however, we observe characteristic audio distortions. We demonstrate that such artefacts are introduced to the vocoder reconstruction by over-smooth mel-spectrogram predictions, which are induced by the choice of mean-squared-error (MSE) loss for training the mel-spectrogram decoder. With MSE loss FastSpeech 2 is limited to learn conditional averages of the training distribution, which might not lie close to a natural sample if the distribution still appears multimodal after all conditioning signals. To alleviate this problem, we introduce TVC-GMM, a mixture model of Trivariate-Chain Gaussian distributions, to model the residual multimodality. TVC-GMM reduces spectrogram smoothness and improves perceptual audio quality in particular for expressive datasets as shown by both objective and subjective evaluation.
comment: Accepted at INTERSPEECH 2023
☆ Interpretable and Explainable Logical Policies via Neurally Guided Symbolic Abstraction
The limited priors required by neural networks make them the dominating choice to encode and learn policies using reinforcement learning (RL). However, they are also black-boxes, making it hard to understand the agent's behaviour, especially when working on the image level. Therefore, neuro-symbolic RL aims at creating policies that are interpretable in the first place. Unfortunately, interpretability is not explainability. To achieve both, we introduce Neurally gUided Differentiable loGic policiEs (NUDGE). NUDGE exploits trained neural network-based agents to guide the search of candidate-weighted logic rules, then uses differentiable logic to train the logic agents. Our experimental evaluation demonstrates that NUDGE agents can induce interpretable and explainable policies while outperforming purely neural ones and showing good flexibility to environments of different initial states and problem sizes.
comment: 9 main pages + appendix (19 in total)
☆ Multi-Objective Population Based Training
Population Based Training (PBT) is an efficient hyperparameter optimization algorithm. PBT is a single-objective algorithm, but many real-world hyperparameter optimization problems involve two or more conflicting objectives. In this work, we therefore introduce a multi-objective version of PBT, MO-PBT. Our experiments on diverse multi-objective hyperparameter optimization problems (Precision/Recall, Accuracy/Fairness, Accuracy/Adversarial Robustness) show that MO-PBT outperforms random search, single-objective PBT, and the state-of-the-art multi-objective hyperparameter optimization algorithm MO-ASHA.
☆ Improving Adversarial Robustness of DEQs with Explicit Regulations Along the Neural Dynamics ICML 2023
Deep equilibrium (DEQ) models replace the multiple-layer stacking of conventional deep networks with a fixed-point iteration of a single-layer transformation. Having been demonstrated to be competitive in a variety of real-world scenarios, the adversarial robustness of general DEQs becomes increasingly crucial for their reliable deployment. Existing works improve the robustness of general DEQ models with the widely-used adversarial training (AT) framework, but they fail to exploit the structural uniquenesses of DEQ models. To this end, we interpret DEQs through the lens of neural dynamics and find that AT under-regulates intermediate states. Besides, the intermediate states typically provide predictions with a high prediction entropy. Informed by the correlation between the entropy of dynamical systems and their stability properties, we propose reducing prediction entropy by progressively updating inputs along the neural dynamics. During AT, we also utilize random intermediate states to compute the loss function. Our methods regulate the neural dynamics of DEQ models in this manner. Extensive experiments demonstrate that our methods substantially increase the robustness of DEQ models and even outperform the strong deep network baselines.
comment: Accepted at ICML 2023. Our code is available at https://github.com/minicheshire/DEQ-Regulating-Neural-Dynamics
☆ Zero-Shot Blind Audio Bandwidth Extension
Audio bandwidth extension involves the realistic reconstruction of high-frequency spectra from bandlimited observations. In cases where the lowpass degradation is unknown, such as in restoring historical audio recordings, this becomes a blind problem. This paper introduces a novel method called BABE (Blind Audio Bandwidth Extension) that addresses the blind problem in a zero-shot setting, leveraging the generative priors of a pre-trained unconditional diffusion model. During the inference process, BABE utilizes a generalized version of diffusion posterior sampling, where the degradation operator is unknown but parametrized and inferred iteratively. The performance of the proposed method is evaluated using objective and subjective metrics, and the results show that BABE surpasses state-of-the-art blind bandwidth extension baselines and achieves competitive performance compared to non-blind filter-informed methods when tested with synthetic data. Moreover, BABE exhibits robust generalization capabilities when enhancing real historical recordings, effectively reconstructing the missing high-frequency content while maintaining coherence with the original recording. Subjective preference tests confirm that BABE significantly improves the audio quality of historical music recordings. Examples of historical recordings restored with the proposed method are available on the companion webpage: (http://research.spa.aalto.fi/publications/papers/ieee-taslp-babe/)
comment: Submitted to IEEE/ACM Transactions on Audio, Speech and Language Processing
☆ Audio-Visual Speech Enhancement with Score-Based Generative Models
This paper introduces an audio-visual speech enhancement system that leverages score-based generative models, also known as diffusion models, conditioned on visual information. In particular, we exploit audio-visual embeddings obtained from a self-super\-vised learning model that has been fine-tuned on lipreading. The layer-wise features of its transformer-based encoder are aggregated, time-aligned, and incorporated into the noise conditional score network. Experimental evaluations show that the proposed audio-visual speech enhancement system yields improved speech quality and reduces generative artifacts such as phonetic confusions with respect to the audio-only equivalent. The latter is supported by the word error rate of a downstream automatic speech recognition model, which decreases noticeably, especially at low input signal-to-noise ratios.
comment: Submitted to ITG Conference on Speech Communication
☆ On Knowledge Editing in Federated Learning: Perspectives, Challenges, and Future Directions
As Federated Learning (FL) has gained increasing attention, it has become widely acknowledged that straightforwardly applying stochastic gradient descent (SGD) on the overall framework when learning over a sequence of tasks results in the phenomenon known as ``catastrophic forgetting''. Consequently, much FL research has centered on devising federated increasing learning methods to alleviate forgetting while augmenting knowledge. On the other hand, forgetting is not always detrimental. The selective amnesia, also known as federated unlearning, which entails the elimination of specific knowledge, can address privacy concerns and create additional ``space'' for acquiring new knowledge. However, there is a scarcity of extensive surveys that encompass recent advancements and provide a thorough examination of this issue. In this manuscript, we present an extensive survey on the topic of knowledge editing (augmentation/removal) in Federated Learning, with the goal of summarizing the state-of-the-art research and expanding the perspective for various domains. Initially, we introduce an integrated paradigm, referred to as Federated Editable Learning (FEL), by reevaluating the entire lifecycle of FL. Secondly, we provide a comprehensive overview of existing methods, evaluate their position within the proposed paradigm, and emphasize the current challenges they face. Lastly, we explore potential avenues for future research and identify unresolved issues.
comment: 7 pages, 1 figure, 2 tabels
☆ A Closer Look at the Adversarial Robustness of Deep Equilibrium Models NeurIPS 2022
Deep equilibrium models (DEQs) refrain from the traditional layer-stacking paradigm and turn to find the fixed point of a single layer. DEQs have achieved promising performance on different applications with featured memory efficiency. At the same time, the adversarial vulnerability of DEQs raises concerns. Several works propose to certify robustness for monotone DEQs. However, limited efforts are devoted to studying empirical robustness for general DEQs. To this end, we observe that an adversarially trained DEQ requires more forward steps to arrive at the equilibrium state, or even violates its fixed-point structure. Besides, the forward and backward tracks of DEQs are misaligned due to the black-box solvers. These facts cause gradient obfuscation when applying the ready-made attacks to evaluate or adversarially train DEQs. Given this, we develop approaches to estimate the intermediate gradients of DEQs and integrate them into the attacking pipelines. Our approaches facilitate fully white-box evaluations and lead to effective adversarial defense for DEQs. Extensive experiments on CIFAR-10 validate the adversarial robustness of DEQs competitive with deep networks of similar sizes.
comment: Accepted at NeurIPS 2022. Our code is available at https://github.com/minicheshire/DEQ-White-Box-Robustness
☆ Improved DeepFake Detection Using Whisper Features INTERSPEECH 2023
With a recent influx of voice generation methods, the threat introduced by audio DeepFake (DF) is ever-increasing. Several different detection methods have been presented as a countermeasure. Many methods are based on so-called front-ends, which, by transforming the raw audio, emphasize features crucial for assessing the genuineness of the audio sample. Our contribution contains investigating the influence of the state-of-the-art Whisper automatic speech recognition model as a DF detection front-end. We compare various combinations of Whisper and well-established front-ends by training 3 detection models (LCNN, SpecRNet, and MesoNet) on a widely used ASVspoof 2021 DF dataset and later evaluating them on the DF In-The-Wild dataset. We show that using Whisper-based features improves the detection for each model and outperforms recent results on the In-The-Wild dataset by reducing Equal Error Rate by 21%.
comment: Accepted to INTERSPEECH 2023
☆ Partial Counterfactual Identification of Continuous Outcomes with a Curvature Sensitivity Model
Counterfactual inference aims to answer retrospective ''what if'' questions and thus belongs to the most fine-grained type of inference in Pearl's causality ladder. Existing methods for counterfactual inference with continuous outcomes aim at point identification and thus make strong and unnatural assumptions about the underlying structural causal model. In this paper, we relax these assumptions and aim at partial counterfactual identification of continuous outcomes, i.e., when the counterfactual query resides in an ignorance interval with informative bounds. We prove that, in general, the ignorance interval of the counterfactual queries has non-informative bounds, already when functions of structural causal models are continuously differentiable. As a remedy, we propose a novel sensitivity model called Curvature Sensitivity Model. This allows us to obtain informative bounds by bounding the curvature of level sets of the functions. We further show that existing point counterfactual identification methods are special cases of our Curvature Sensitivity Model when the bound of the curvature is set to zero. We then propose an implementation of our Curvature Sensitivity Model in the form of a novel deep generative model, which we call Augmented Pseudo-Invertible Decoder. Our implementation employs (i) residual normalizing flows with (ii) variational augmentations. We empirically demonstrate the effectiveness of our Augmented Pseudo-Invertible Decoder. To the best of our knowledge, ours is the first partial identification model for Markovian structural causal models with continuous outcomes.
☆ Leveraging the Triple Exponential Moving Average for Fast-Adaptive Moment Estimation
Network optimization is a crucial step in the field of deep learning, as it directly affects the performance of models in various domains such as computer vision. Despite the numerous optimizers that have been developed over the years, the current methods are still limited in their ability to accurately and quickly identify gradient trends, which can lead to sub-optimal network performance. In this paper, we propose a novel deep optimizer called Fast-Adaptive Moment Estimation (FAME), which for the first time estimates gradient moments using a Triple Exponential Moving Average (TEMA). Incorporating TEMA into the optimization process provides richer and more accurate information on data changes and trends, as compared to the standard Exponential Moving Average used in essentially all current leading adaptive optimization methods. Our proposed FAME optimizer has been extensively validated through a wide range of benchmarks, including CIFAR-10, CIFAR-100, PASCAL-VOC, MS-COCO, and Cityscapes, using 14 different learning architectures, six optimizers, and various vision tasks, including detection, classification and semantic understanding. The results demonstrate that our FAME optimizer outperforms other leading optimizers in terms of both robustness and accuracy.
☆ The Flawed Foundations of Fair Machine Learning
The definition and implementation of fairness in automated decisions has been extensively studied by the research community. Yet, there hides fallacious reasoning, misleading assertions, and questionable practices at the foundations of the current fair machine learning paradigm. Those flaws are the result of a failure to understand that the trade-off between statistically accurate outcomes and group similar outcomes exists as independent, external constraint rather than as a subjective manifestation as has been commonly argued. First, we explain that there is only one conception of fairness present in the fair machine learning literature: group similarity of outcomes based on a sensitive attribute where the similarity benefits an underprivileged group. Second, we show that there is, in fact, a trade-off between statistically accurate outcomes and group similar outcomes in any data setting where group disparities exist, and that the trade-off presents an existential threat to the equitable, fair machine learning approach. Third, we introduce a proof-of-concept evaluation to aid researchers and designers in understanding the relationship between statistically accurate outcomes and group similar outcomes. Finally, suggestions for future work aimed at data scientists, legal scholars, and data ethicists that utilize the conceptual and experimental framework described throughout this article are provided.
comment: This article is a preprint submitted to the Minds and Machines Special Issue on the (Un)fairness of AI on May 31st, 2023
☆ Adaptive Attractors: A Defense Strategy against ML Adversarial Collusion Attacks
In the seller-buyer setting on machine learning models, the seller generates different copies based on the original model and distributes them to different buyers, such that adversarial samples generated on one buyer's copy would likely not work on other copies. A known approach achieves this using attractor-based rewriter which injects different attractors to different copies. This induces different adversarial regions in different copies, making adversarial samples generated on one copy not replicable on others. In this paper, we focus on a scenario where multiple malicious buyers collude to attack. We first give two formulations and conduct empirical studies to analyze effectiveness of collusion attack under different assumptions on the attacker's capabilities and properties of the attractors. We observe that existing attractor-based methods do not effectively mislead the colluders in the sense that adversarial samples found are influenced more by the original model instead of the attractors as number of colluders increases. Based on this observation, we propose using adaptive attractors whose weight is guided by a U-shape curve to cover the shortfalls. Experimentation results show that when using our approach, the attack success rate of a collusion attack converges to around 15% even when lots of copies are applied for collusion. In contrast, when using the existing attractor-based rewriter with fixed weight, the attack success rate increases linearly with the number of copies used for collusion.
☆ Chemical Property-Guided Neural Networks for Naphtha Composition Prediction
The naphtha cracking process heavily relies on the composition of naphtha, which is a complex blend of different hydrocarbons. Predicting the naphtha composition accurately is crucial for efficiently controlling the cracking process and achieving maximum performance. Traditional methods, such as gas chromatography and true boiling curve, are not feasible due to the need for pilot-plant-scale experiments or cost constraints. In this paper, we propose a neural network framework that utilizes chemical property information to improve the performance of naphtha composition prediction. Our proposed framework comprises two parts: a Watson K factor estimation network and a naphtha composition prediction network. Both networks share a feature extraction network based on Convolutional Neural Network (CNN) architecture, while the output layers use Multi-Layer Perceptron (MLP) based networks to generate two different outputs - Watson K factor and naphtha composition. The naphtha composition is expressed in percentages, and its sum should be 100%. To enhance the naphtha composition prediction, we utilize a distillation simulator to obtain the distillation curve from the naphtha composition, which is dependent on its chemical properties. By designing a loss function between the estimated and simulated Watson K factors, we improve the performance of both Watson K estimation and naphtha composition prediction. The experimental results show that our proposed framework can predict the naphtha composition accurately while reflecting real naphtha chemical properties.
comment: Accepted at IEEE International Conference on Industrial Informatics 2023(INDIN 2023)
☆ Adaptive Message Quantization and Parallelization for Distributed Full-graph GNN Training
Distributed full-graph training of Graph Neural Networks (GNNs) over large graphs is bandwidth-demanding and time-consuming. Frequent exchanges of node features, embeddings and embedding gradients (all referred to as messages) across devices bring significant communication overhead for nodes with remote neighbors on other devices (marginal nodes) and unnecessary waiting time for nodes without remote neighbors (central nodes) in the training graph. This paper proposes an efficient GNN training system, AdaQP, to expedite distributed full-graph GNN training. We stochastically quantize messages transferred across devices to lower-precision integers for communication traffic reduction and advocate communication-computation parallelization between marginal nodes and central nodes. We provide theoretical analysis to prove fast training convergence (at the rate of O(T^{-1}) with T being the total number of training epochs) and design an adaptive quantization bit-width assignment scheme for each message based on the analysis, targeting a good trade-off between training convergence and efficiency. Extensive experiments on mainstream graph datasets show that AdaQP substantially improves distributed full-graph training's throughput (up to 3.01 X) with negligible accuracy drop (at most 0.30%) or even accuracy improvement (up to 0.19%) in most cases, showing significant advantages over the state-of-the-art works.
☆ DSHGT: Dual-Supervisors Heterogeneous Graph Transformer -- A pioneer study of using heterogeneous graph learning for detecting software vulnerabilities
Vulnerability detection is a critical problem in software security and attracts growing attention both from academia and industry. Traditionally, software security is safeguarded by designated rule-based detectors that heavily rely on empirical expertise, requiring tremendous effort from software experts to generate rule repositories for large code corpus. Recent advances in deep learning, especially Graph Neural Networks (GNN), have uncovered the feasibility of automatic detection of a wide range of software vulnerabilities. However, prior learning-based works only break programs down into a sequence of word tokens for extracting contextual features of codes, or apply GNN largely on homogeneous graph representation (e.g., AST) without discerning complex types of underlying program entities (e.g., methods, variables). In this work, we are one of the first to explore heterogeneous graph representation in the form of Code Property Graph and adapt a well-known heterogeneous graph network with a dual-supervisor structure for the corresponding graph learning task. Using the prototype built, we have conducted extensive experiments on both synthetic datasets and real-world projects. Compared with the state-of-the-art baselines, the results demonstrate promising effectiveness in this research direction in terms of vulnerability detection performance (average F1 improvements over 10\% in real-world projects) and transferability from C/C++ to other programming languages (average F1 improvements over 11%).
☆ Robust and Generalisable Segmentation of Subtle Epilepsy-causing Lesions: a Graph Convolutional Approach MICCAI 2023
Focal cortical dysplasia (FCD) is a leading cause of drug-resistant focal epilepsy, which can be cured by surgery. These lesions are extremely subtle and often missed even by expert neuroradiologists. "Ground truth" manual lesion masks are therefore expensive, limited and have large inter-rater variability. Existing FCD detection methods are limited by high numbers of false positive predictions, primarily due to vertex- or patch-based approaches that lack whole-brain context. Here, we propose to approach the problem as semantic segmentation using graph convolutional networks (GCN), which allows our model to learn spatial relationships between brain regions. To address the specific challenges of FCD identification, our proposed model includes an auxiliary loss to predict distance from the lesion to reduce false positives and a weak supervision classification loss to facilitate learning from uncertain lesion masks. On a multi-centre dataset of 1015 participants with surface-based features and manual lesion masks from structural MRI data, the proposed GCN achieved an AUC of 0.74, a significant improvement against a previously used vertex-wise multi-layer perceptron (MLP) classifier (AUC 0.64). With sensitivity thresholded at 67%, the GCN had a specificity of 71% in comparison to 49% when using the MLP. This improvement in specificity is vital for clinical integration of lesion-detection tools into the radiological workflow, through increasing clinical confidence in the use of AI radiological adjuncts and reducing the number of areas requiring expert review.
comment: accepted at MICCAI 2023
☆ DWT-CompCNN: Deep Image Classification Network for High Throughput JPEG 2000 Compressed Documents
For any digital application with document images such as retrieval, the classification of document images becomes an essential stage. Conventionally for the purpose, the full versions of the documents, that is the uncompressed document images make the input dataset, which poses a threat due to the big volume required to accommodate the full versions of the documents. Therefore, it would be novel, if the same classification task could be accomplished directly (with some partial decompression) with the compressed representation of documents in order to make the whole process computationally more efficient. In this research work, a novel deep learning model, DWT CompCNN is proposed for classification of documents that are compressed using High Throughput JPEG 2000 (HTJ2K) algorithm. The proposed DWT-CompCNN comprises of five convolutional layers with filter sizes of 16, 32, 64, 128, and 256 consecutively for each increasing layer to improve learning from the wavelet coefficients extracted from the compressed images. Experiments are performed on two benchmark datasets- Tobacco-3482 and RVL-CDIP, which demonstrate that the proposed model is time and space efficient, and also achieves a better classification accuracy in compressed domain.
comment: In Springer Journal - Pattern Analysis and Applications under Minor Revision
☆ Covert Communication Based on the Poisoning Attack in Federated Learning
Covert communication has become an important area of research in computer security. It involves hiding specific information on a carrier for message transmission and is often used to transmit private data, military secrets, and even malware. In deep learning, many methods have been developed for hiding information in models to achieve covert communication. However, these methods are not applicable to federated learning, where model aggregation invalidates the exact information embedded in the model by the client. To address this problem, we propose a novel method for covert communication in federated learning based on the poisoning attack. Our approach achieves 100% accuracy in covert message transmission between two clients and is shown to be both stealthy and robust through extensive experiments. However, existing defense methods are limited in their effectiveness against our attack scheme, highlighting the urgent need for new protection methods to be developed. Our study emphasizes the necessity of research in covert communication and serves as a foundation for future research in federated learning attacks and defenses.
☆ Resource-Efficient Federated Hyperdimensional Computing
In conventional federated hyperdimensional computing (HDC), training larger models usually results in higher predictive performance but also requires more computational, communication, and energy resources. If the system resources are limited, one may have to sacrifice the predictive performance by reducing the size of the HDC model. The proposed resource-efficient federated hyperdimensional computing (RE-FHDC) framework alleviates such constraints by training multiple smaller independent HDC sub-models and refining the concatenated HDC model using the proposed dropout-inspired procedure. Our numerical comparison demonstrates that the proposed framework achieves a comparable or higher predictive performance while consuming less computational and wireless resources than the baseline federated HDC implementation.
comment: Accepted to Federated Learning Systems (FLSys) workshop, in Conjunction with the 6th MLSys Conference (MLSys 2023)
☆ Federated Domain Generalization: A Survey
Machine learning typically relies on the assumption that training and testing distributions are identical and that data is centrally stored for training and testing. However, in real-world scenarios, distributions may differ significantly and data is often distributed across different devices, organizations, or edge nodes. Consequently, it is imperative to develop models that can effectively generalize to unseen distributions where data is distributed across different domains. In response to this challenge, there has been a surge of interest in federated domain generalization (FDG) in recent years. FDG combines the strengths of federated learning (FL) and domain generalization (DG) techniques to enable multiple source domains to collaboratively learn a model capable of directly generalizing to unseen domains while preserving data privacy. However, generalizing the federated model under domain shifts is a technically challenging problem that has received scant attention in the research area so far. This paper presents the first survey of recent advances in this area. Initially, we discuss the development process from traditional machine learning to domain adaptation and domain generalization, leading to FDG as well as provide the corresponding formal definition. Then, we categorize recent methodologies into four classes: federated domain alignment, data manipulation, learning strategies, and aggregation optimization, and present suitable algorithms in detail for each category. Next, we introduce commonly used datasets, applications, evaluations, and benchmarks. Finally, we conclude this survey by providing some potential research topics for the future.
☆ Navigating Fairness in Radiology AI: Concepts, Consequences,and Crucial Considerations
Artificial Intelligence (AI) has significantly revolutionized radiology, promising improved patient outcomes and streamlined processes. However, it's critical to ensure the fairness of AI models to prevent stealthy bias and disparities from leading to unequal outcomes. This review discusses the concept of fairness in AI, focusing on bias auditing using the Aequitas toolkit, and its real-world implications in radiology, particularly in disease screening scenarios. Aequitas, an open-source bias audit toolkit, scrutinizes AI models' decisions, identifying hidden biases that may result in disparities across different demographic groups and imaging equipment brands. This toolkit operates on statistical theories, analyzing a large dataset to reveal a model's fairness. It excels in its versatility to handle various variables simultaneously, especially in a field as diverse as radiology. The review explicates essential fairness metrics: Equal and Proportional Parity, False Positive Rate Parity, False Discovery Rate Parity, False Negative Rate Parity, and False Omission Rate Parity. Each metric serves unique purposes and offers different insights. We present hypothetical scenarios to demonstrate their relevance in disease screening settings, and how disparities can lead to significant real-world impacts.
☆ Differentiable Grey-box Modelling of Phaser Effects using Frame-based Spectral Processing
Machine learning approaches to modelling analog audio effects have seen intensive investigation in recent years, particularly in the context of non-linear time-invariant effects such as guitar amplifiers. For modulation effects such as phasers, however, new challenges emerge due to the presence of the low-frequency oscillator which controls the slowly time-varying nature of the effect. Existing approaches have either required foreknowledge of this control signal, or have been non-causal in implementation. This work presents a differentiable digital signal processing approach to modelling phaser effects in which the underlying control signal and time-varying spectral response of the effect are jointly learned. The proposed model processes audio in short frames to implement a time-varying filter in the frequency domain, with a transfer function based on typical analog phaser circuit topology. We show that the model can be trained to emulate an analog reference device, while retaining interpretable and adjustable parameters. The frame duration is an important hyper-parameter of the proposed model, so an investigation was carried out into its effect on model accuracy. The optimal frame length depends on both the rate and transient decay-time of the target effect, but the frame length can be altered at inference time without a significant change in accuracy.
comment: Accepted for publication in Proc. DAFx23, Copenhagen, Denmark, September 2023
☆ Hyperparameters in Reinforcement Learning and How To Tune Them
In order to improve reproducibility, deep reinforcement learning (RL) has been adopting better scientific practices such as standardized evaluation metrics and reporting. However, the process of hyperparameter optimization still varies widely across papers, which makes it challenging to compare RL algorithms fairly. In this paper, we show that hyperparameter choices in RL can significantly affect the agent's final performance and sample efficiency, and that the hyperparameter landscape can strongly depend on the tuning seed which may lead to overfitting. We therefore propose adopting established best practices from AutoML, such as the separation of tuning and testing seeds, as well as principled hyperparameter optimization (HPO) across a broad search space. We support this by comparing multiple state-of-the-art HPO tools on a range of RL algorithms and environments to their hand-tuned counterparts, demonstrating that HPO approaches often have higher performance and lower compute overhead. As a result of our findings, we recommend a set of best practices for the RL community, which should result in stronger empirical results with fewer computational costs, better reproducibility, and thus faster progress. In order to encourage the adoption of these practices, we provide plug-and-play implementations of the tuning algorithms used in this paper at https://github.com/facebookresearch/how-to-autorl.
♻ ☆ Scaling in Depth: Unlocking Robustness Certification on ImageNet
Despite the promise of Lipschitz-based methods for provably-robust deep learning with deterministic guarantees, current state-of-the-art results are limited to feed-forward Convolutional Networks (ConvNets) on low-dimensional data, such as CIFAR-10. This paper investigates strategies for expanding certifiably robust training to larger, deeper models. A key challenge in certifying deep networks is efficient calculation of the Lipschitz bound for residual blocks found in ResNet and ViT architectures. We show that fast ways of bounding the Lipschitz constant for conventional ResNets are loose, and show how to address this by designing a new residual block, leading to the \emph{Linear ResNet} (LiResNet) architecture. We then introduce \emph{Efficient Margin MAximization} (EMMA), a loss function that stabilizes robust training by simultaneously penalizing worst-case adversarial examples from \emph{all} classes. Together, these contributions yield new \emph{state-of-the-art} robust accuracy on CIFAR-10/100 and Tiny-ImageNet under $\ell_2$ perturbations. Moreover, for the first time, we are able to scale up fast deterministic robustness guarantees to ImageNet, demonstrating that this approach to robust learning can be applied to real-world applications. We release our code on Github: \url{https://github.com/klasleino/gloro}.
♻ ☆ Online Continuous Hyperparameter Optimization for Contextual Bandits
In stochastic contextual bandits, an agent sequentially makes actions from a time-dependent action set based on past experience to minimize the cumulative regret. Like many other machine learning algorithms, the performance of bandits heavily depends on their multiple hyperparameters, and theoretically derived parameter values may lead to unsatisfactory results in practice. Moreover, it is infeasible to use offline tuning methods like cross-validation to choose hyperparameters under the bandit environment, as the decisions should be made in real time. To address this challenge, we propose the first online continuous hyperparameter tuning framework for contextual bandits to learn the optimal parameter configuration within a search space on the fly. Specifically, we use a double-layer bandit framework named CDT (Continuous Dynamic Tuning) and formulate the hyperparameter optimization as a non-stationary continuum-armed bandit, where each arm represents a combination of hyperparameters, and the corresponding reward is the algorithmic result. For the top layer, we propose the Zooming TS algorithm that utilizes Thompson Sampling (TS) for exploration and a restart technique to get around the switching environment. The proposed CDT framework can be easily used to tune contextual bandit algorithms without any pre-specified candidate set for hyperparameters. We further show that it could achieve sublinear regret in theory and performs consistently better on both synthetic and real datasets in practice.
♻ ☆ Diffusion Self-Guidance for Controllable Image Generation
Large-scale generative models are capable of producing high-quality images from detailed text descriptions. However, many aspects of an image are difficult or impossible to convey through text. We introduce self-guidance, a method that provides greater control over generated images by guiding the internal representations of diffusion models. We demonstrate that properties such as the shape, location, and appearance of objects can be extracted from these representations and used to steer sampling. Self-guidance works similarly to classifier guidance, but uses signals present in the pretrained model itself, requiring no additional models or training. We show how a simple set of properties can be composed to perform challenging image manipulations, such as modifying the position or size of objects, merging the appearance of objects in one image with the layout of another, composing objects from many images into one, and more. We also show that self-guidance can be used to edit real images. For results and an interactive demo, see our project page at https://dave.ml/selfguidance/
comment: Project page at https://dave.ml/selfguidance/
♻ ☆ Enhanced Gaussian Process Dynamical Models with Knowledge Transfer for Long-term Battery Degradation Forecasting
Predicting the end-of-life or remaining useful life of batteries in electric vehicles is a critical and challenging problem, predominantly approached in recent years using machine learning to predict the evolution of the state-of-health during repeated cycling. To improve the accuracy of predictive estimates, especially early in the battery lifetime, a number of algorithms have incorporated features that are available from data collected by battery management systems. Unless multiple battery data sets are used for a direct prediction of the end-of-life, which is useful for ball-park estimates, such an approach is infeasible since the features are not known for future cycles. In this paper, we develop a highly-accurate method that can overcome this limitation, by using a modified Gaussian process dynamical model (GPDM). We introduce a kernelised version of GPDM for a more expressive covariance structure between both the observable and latent coordinates. We combine the approach with transfer learning to track the future state-of-health up to end-of-life. The method can incorporate features as different physical observables, without requiring their values beyond the time up to which data is available. Transfer learning is used to improve learning of the hyperparameters using data from similar batteries. The accuracy and superiority of the approach over modern benchmarks algorithms including a Gaussian process model and deep convolutional and recurrent networks are demonstrated on three data sets, particularly at the early stages of the battery lifetime.
♻ ☆ PrefRec: Recommender Systems with Human Preferences for Reinforcing Long-term User Engagement
Current advances in recommender systems have been remarkably successful in optimizing immediate engagement. However, long-term user engagement, a more desirable performance metric, remains difficult to improve. Meanwhile, recent reinforcement learning (RL) algorithms have shown their effectiveness in a variety of long-term goal optimization tasks. For this reason, RL is widely considered as a promising framework for optimizing long-term user engagement in recommendation. Though promising, the application of RL heavily relies on well-designed rewards, but designing rewards related to long-term user engagement is quite difficult. To mitigate the problem, we propose a novel paradigm, recommender systems with human preferences (or Preference-based Recommender systems), which allows RL recommender systems to learn from preferences about users historical behaviors rather than explicitly defined rewards. Such preferences are easily accessible through techniques such as crowdsourcing, as they do not require any expert knowledge. With PrefRec, we can fully exploit the advantages of RL in optimizing long-term goals, while avoiding complex reward engineering. PrefRec uses the preferences to automatically train a reward function in an end-to-end manner. The reward function is then used to generate learning signals to train the recommendation policy. Furthermore, we design an effective optimization method for PrefRec, which uses an additional value function, expectile regression and reward model pre-training to improve the performance. We conduct experiments on a variety of long-term user engagement optimization tasks. The results show that PrefRec significantly outperforms previous state-of-the-art methods in all the tasks.
♻ ☆ Contextualize Me -- The Case for Context in Reinforcement Learning
While Reinforcement Learning ( RL) has made great strides towards solving increasingly complicated problems, many algorithms are still brittle to even slight environmental changes. Contextual Reinforcement Learning (cRL) provides a framework to model such changes in a principled manner, thereby enabling flexible, precise and interpretable task specification and generation. Our goal is to show how the framework of cRL contributes to improving zero-shot generalization in RL through meaningful benchmarks and structured reasoning about generalization tasks. We confirm the insight that optimal behavior in cRL requires context information, as in other related areas of partial observability. To empirically validate this in the cRL framework, we provide various context-extended versions of common RL environments. They are part of the first benchmark library, CARL, designed for generalization based on cRL extensions of popular benchmarks, which we propose as a testbed to further study general agents. We show that in the contextual setting, even simple RL environments become challenging - and that naive solutions are not enough to generalize across complex context spaces.
comment: arXiv admin note: substantial text overlap with arXiv:2110.02102
♻ ☆ Efficient Latency-Aware CNN Depth Compression via Two-Stage Dynamic Programming ICML 2023
Recent works on neural network pruning advocate that reducing the depth of the network is more effective in reducing run-time memory usage and accelerating inference latency than reducing the width of the network through channel pruning. In this regard, some recent works propose depth compression algorithms that merge convolution layers. However, the existing algorithms have a constricted search space and rely on human-engineered heuristics. In this paper, we propose a novel depth compression algorithm which targets general convolution operations. We propose a subset selection problem that replaces inefficient activation layers with identity functions and optimally merges consecutive convolution operations into shallow equivalent convolution operations for efficient end-to-end inference latency. Since the proposed subset selection problem is NP-hard, we formulate a surrogate optimization problem that can be solved exactly via two-stage dynamic programming within a few seconds. We evaluate our methods and baselines by TensorRT for a fair inference latency comparison. Our method outperforms the baseline method with higher accuracy and faster inference speed in MobileNetV2 on the ImageNet dataset. Specifically, we achieve $1.41\times$ speed-up with $0.11$\%p accuracy gain in MobileNetV2-1.0 on the ImageNet.
comment: ICML 2023; Codes at https://github.com/snu-mllab/Efficient-CNN-Depth-Compression
♻ ☆ Visual Question Answering: A Survey on Techniques and Common Trends in Recent Literature
Visual Question Answering (VQA) is an emerging area of interest for researches, being a recent problem in natural language processing and image prediction. In this area, an algorithm needs to answer questions about certain images. As of the writing of this survey, 25 recent studies were analyzed. Besides, 6 datasets were analyzed and provided their link to download. In this work, several recent pieces of research in this area were investigated and a deeper analysis and comparison among them were provided, including results, the state-of-the-art, common errors, and possible points of improvement for future researchers.
comment: 30 pages. arXiv admin note: text overlap with arXiv:2104.00926, arXiv:2110.02526, arXiv:2108.02059, arXiv:1908.01801 by other authors
♻ ☆ Tighter Information-Theoretic Generalization Bounds from Supersamples ICML 2023
In this work, we present a variety of novel information-theoretic generalization bounds for learning algorithms, from the supersample setting of Steinke & Zakynthinou (2020)-the setting of the "conditional mutual information" framework. Our development exploits projecting the loss pair (obtained from a training instance and a testing instance) down to a single number and correlating loss values with a Rademacher sequence (and its shifted variants). The presented bounds include square-root bounds, fast-rate bounds, including those based on variance and sharpness, and bounds for interpolating algorithms etc. We show theoretically or empirically that these bounds are tighter than all information-theoretic bounds known to date on the same supersample setting.
comment: Accepted to ICML 2023
♻ ☆ Consistent and fast inference in compartmental models of epidemics using Poisson Approximate Likelihoods
Addressing the challenge of scaling-up epidemiological inference to complex and heterogeneous models, we introduce Poisson Approximate Likelihood (PAL) methods. In contrast to the popular ODE approach to compartmental modelling, in which a large population limit is used to motivate a deterministic model, PALs are derived from approximate filtering equations for finite-population, stochastic compartmental models, and the large population limit drives consistency of maximum PAL estimators. Our theoretical results appear to be the first likelihood-based parameter estimation consistency results which apply to a broad class of partially observed stochastic compartmental models and address the large population limit. PALs are simple to implement, involving only elementary arithmetic operations and no tuning parameters, and fast to evaluate, requiring no simulation from the model and having computational cost independent of population size. Through examples we demonstrate how PALs can be used to: fit an age-structured model of influenza, taking advantage of automatic differentiation in Stan; compare over-dispersion mechanisms in a model of rotavirus by embedding PALs within sequential Monte Carlo; and evaluate the role of unit-specific parameters in a meta-population model of measles.
♻ ☆ Supervised Metric Learning to Rank for Retrieval via Contextual Similarity Optimization
There is extensive interest in metric learning methods for image retrieval. Many metric learning loss functions focus on learning a correct ranking of training samples, but strongly overfit semantically inconsistent labels and require a large amount of data. To address these shortcomings, we propose a new metric learning method, called contextual loss, which optimizes contextual similarity in addition to cosine similarity. Our contextual loss implicitly enforces semantic consistency among neighbors while converging to the correct ranking. We empirically show that the proposed loss is more robust to label noise, and is less prone to overfitting even when a large portion of train data is withheld. Extensive experiments demonstrate that our method achieves a new state-of-the-art across four image retrieval benchmarks and multiple different evaluation settings. Code is available at: https://github.com/Chris210634/metric-learning-using-contextual-similarity
♻ ☆ Are Deep Neural Networks SMARTer than Second Graders? CVPR 2023
Recent times have witnessed an increasing number of applications of deep neural networks towards solving tasks that require superior cognitive abilities, e.g., playing Go, generating art, ChatGPT, etc. Such a dramatic progress raises the question: how generalizable are neural networks in solving problems that demand broad skills? To answer this question, we propose SMART: a Simple Multimodal Algorithmic Reasoning Task and the associated SMART-101 dataset, for evaluating the abstraction, deduction, and generalization abilities of neural networks in solving visuo-linguistic puzzles designed specifically for children in the 6--8 age group. Our dataset consists of 101 unique puzzles; each puzzle comprises a picture and a question, and their solution needs a mix of several elementary skills, including arithmetic, algebra, and spatial reasoning, among others. To scale our dataset towards training deep neural networks, we programmatically generate entirely new instances for each puzzle, while retaining their solution algorithm. To benchmark performances on SMART-101, we propose a vision and language meta-learning model using varied state-of-the-art backbones. Our experiments reveal that while powerful deep models offer reasonable performances on puzzles in a supervised setting, they are not better than random accuracy when analyzed for generalization. We also evaluate the recent ChatGPT and other large language models on a part of SMART-101 and find that while these models show convincing reasoning abilities, the answers are often incorrect.
comment: Accepted at CVPR 2023. For the SMART-101 dataset, see https://doi.org/10.5281/zenodo.7761799
♻ ☆ Neural Wasserstein Gradient Flows for Maximum Mean Discrepancies with Riesz Kernels
Wasserstein gradient flows of maximum mean discrepancy (MMD) functionals with non-smooth Riesz kernels show a rich structure as singular measures can become absolutely continuous ones and conversely. In this paper we contribute to the understanding of such flows. We propose to approximate the backward scheme of Jordan, Kinderlehrer and Otto for computing such Wasserstein gradient flows as well as a forward scheme for so-called Wasserstein steepest descent flows by neural networks (NNs). Since we cannot restrict ourselves to absolutely continuous measures, we have to deal with transport plans and velocity plans instead of usual transport maps and velocity fields. Indeed, we approximate the disintegration of both plans by generative NNs which are learned with respect to appropriate loss functions. In order to evaluate the quality of both neural schemes, we benchmark them on the interaction energy. Here we provide analytic formulas for Wasserstein schemes starting at a Dirac measure and show their convergence as the time step size tends to zero. Finally, we illustrate our neural MMD flows by numerical examples.
comment: arXiv admin note: text overlap with arXiv:2211.01804
♻ ☆ Convergence of the Inexact Langevin Algorithm and Score-based Generative Models in KL Divergence
We study the Inexact Langevin Dynamics (ILD), Inexact Langevin Algorithm (ILA), and Score-based Generative Modeling (SGM) when utilizing estimated score functions for sampling. Our focus lies in establishing stable biased convergence guarantees in terms of the Kullback-Leibler (KL) divergence. To achieve these guarantees, we impose two key assumptions: 1) the target distribution satisfies the log-Sobolev inequality (LSI), and 2) the score estimator exhibits a bounded Moment Generating Function (MGF) error. Notably, the MGF error assumption we adopt is more lenient compared to the $L^\infty$ error assumption used in existing literature. However, it is stronger than the $L^2$ error assumption utilized in recent works, which often leads to unstable bounds. We explore the question of how to obtain a provably accurate score estimator that satisfies the MGF error assumption. Specifically, we demonstrate that a simple estimator based on kernel density estimation fulfills the MGF error assumption for sub-Gaussian target distribution, at the population level.
♻ ☆ Locally Regularized Neural Differential Equations: Some Black Boxes Were Meant to Remain Closed!
Implicit layer deep learning techniques, like Neural Differential Equations, have become an important modeling framework due to their ability to adapt to new problems automatically. Training a neural differential equation is effectively a search over a space of plausible dynamical systems. However, controlling the computational cost for these models is difficult since it relies on the number of steps the adaptive solver takes. Most prior works have used higher-order methods to reduce prediction timings while greatly increasing training time or reducing both training and prediction timings by relying on specific training algorithms, which are harder to use as a drop-in replacement due to strict requirements on automatic differentiation. In this manuscript, we use internal cost heuristics of adaptive differential equation solvers at stochastic time points to guide the training toward learning a dynamical system that is easier to integrate. We "close the black-box" and allow the use of our method with any adjoint technique for gradient calculations of the differential equation solution. We perform experimental studies to compare our method to global regularization to show that we attain similar performance numbers without compromising the flexibility of implementation on ordinary differential equations (ODEs) and stochastic differential equations (SDEs). We develop two sampling strategies to trade off between performance and training time. Our method reduces the number of function evaluations to 0.556-0.733x and accelerates predictions by 1.3-2x.
comment: Proceedings of the 40th International Conference on Machine Learning, Honolulu, Hawaii, USA. PMLR 202, 2023
♻ ☆ A Note On Interpreting Canary Exposure
Canary exposure, introduced in Carlini et al. is frequently used to empirically evaluate, or audit, the privacy of machine learning model training. The goal of this note is to provide some intuition on how to interpret canary exposure, including by relating it to membership inference attacks and differential privacy.
comment: short note, edited to add a sentence on independence of canary losses, including adding Pillutla et al
♻ ☆ Maximum Entropy on Erroneous Predictions (MEEP): Improving model calibration for medical image segmentation MICCAI 2023
Modern deep neural networks achieved remarkable progress in medical image segmentation tasks. However, it has recently been observed that they tend to produce overconfident estimates, even in situations of high uncertainty, leading to poorly calibrated and unreliable models. In this work we introduce Maximum Entropy on Erroneous Predictions (MEEP), a training strategy for segmentation networks which selectively penalizes overconfident predictions, focusing only on misclassified pixels. Our method is agnostic to the neural architecture, does not increase model complexity and can be coupled with multiple segmentation loss functions. We benchmark the proposed strategy in two challenging segmentation tasks: white matter hyperintensity lesions in magnetic resonance images (MRI) of the brain, and atrial segmentation in cardiac MRI. The experimental results demonstrate that coupling MEEP with standard segmentation losses leads to improvements not only in terms of model calibration, but also in segmentation quality.
comment: Accepted for publication at MICCAI 2023
♻ ☆ Towards Understanding the Dynamics of Gaussian-Stein Variational Gradient Descent
Stein Variational Gradient Descent (SVGD) is a nonparametric particle-based deterministic sampling algorithm. Despite its wide usage, understanding the theoretical properties of SVGD has remained a challenging problem. For sampling from a Gaussian target, the SVGD dynamics with a bilinear kernel will remain Gaussian as long as the initializer is Gaussian. Inspired by this fact, we undertake a detailed theoretical study of the Gaussian-SVGD, i.e., SVGD projected to the family of Gaussian distributions via the bilinear kernel, or equivalently Gaussian variational inference (GVI) with SVGD. We present a complete picture by considering both the mean-field PDE and discrete particle systems. When the target is strongly log-concave, the mean-field Gaussian-SVGD dynamics is proven to converge linearly to the Gaussian distribution closest to the target in KL divergence. In the finite-particle setting, there is both uniform in time convergence to the mean-field limit and linear convergence in time to the equilibrium if the target is Gaussian. In the general case, we propose a density-based and a particle-based implementation of the Gaussian-SVGD, and show that several recent algorithms for GVI, proposed from different perspectives, emerge as special cases of our unified framework. Interestingly, one of the new particle-based instance from this framework empirically outperforms existing approaches. Our results make concrete contributions towards obtaining a deeper understanding of both SVGD and GVI.
comment: 59 pages, 7 figures
♻ ☆ Optimal transport flow and infinitesimal density ratio estimation
Continuous normalizing flows are widely used in generative tasks, where a flow network transports from a data distribution $P$ to a normal distribution. A flow model that transports from $P$ to an arbitrary $Q$, where both $P$ and $Q$ are accessible via finite samples, is of various application interests, particularly in the recently developed telescoping density ratio estimation (DRE) which calls for the construction of intermediate densities to bridge between the two densities. In this work, we propose such a flow by a neural-ODE model which is trained from empirical samples to transport invertibly from $P$ to $Q$ (and vice versa) and optimally by minimizing the transport cost. The trained flow model allows us to perform infinitesimal DRE along the time-parametrized $\log$-density by training an additional continuous-time network using classification loss, whose time integration provides a telescopic DRE. The effectiveness of the proposed model is empirically demonstrated on high-dimensional mutual information estimation and energy-based generative models of image data.
♻ ☆ Can Deep Learning Reliably Recognize Abnormality Patterns on Chest X-rays? A Multi-Reader Study Examining One Month of AI Implementation in Everyday Radiology Clinical Practice
In this study, we developed a deep-learning-based automatic detection algorithm (DLAD, Carebot AI CXR) to detect and localize seven specific radiological findings (atelectasis (ATE), consolidation (CON), pleural effusion (EFF), pulmonary lesion (LES), subcutaneous emphysema (SCE), cardiomegaly (CMG), pneumothorax (PNO)) on chest X-rays (CXR). We collected 956 CXRs and compared the performance of the DLAD with that of six individual radiologists who assessed the images in a hospital setting. The proposed DLAD achieved high sensitivity (ATE 1.000 (0.624-1.000), CON 0.864 (0.671-0.956), EFF 0.953 (0.887-0.983), LES 0.905 (0.715-0.978), SCE 1.000 (0.366-1.000), CMG 0.837 (0.711-0.917), PNO 0.875 (0.538-0.986)), even when compared to the radiologists (LOWEST: ATE 0.000 (0.000-0.376), CON 0.182 (0.070-0.382), EFF 0.400 (0.302-0.506), LES 0.238 (0.103-0.448), SCE 0.000 (0.000-0.634), CMG 0.347 (0.228-0.486), PNO 0.375 (0.134-0.691), HIGHEST: ATE 1.000 (0.624-1.000), CON 0.864 (0.671-0.956), EFF 0.953 (0.887-0.983), LES 0.667 (0.456-0.830), SCE 1.000 (0.366-1.000), CMG 0.980 (0.896-0.999), PNO 0.875 (0.538-0.986)). The findings of the study demonstrate that the suggested DLAD holds potential for integration into everyday clinical practice as a decision support system, effectively mitigating the false negative rate associated with junior and intermediate radiologists.
♻ ☆ Matching-based Data Valuation for Generative Model
Data valuation is critical in machine learning, as it helps enhance model transparency and protect data properties. Existing data valuation methods have primarily focused on discriminative models, neglecting deep generative models that have recently gained considerable attention. Similar to discriminative models, there is an urgent need to assess data contributions in deep generative models as well. However, previous data valuation approaches mainly relied on discriminative model performance metrics and required model retraining. Consequently, they cannot be applied directly and efficiently to recent deep generative models, such as generative adversarial networks and diffusion models, in practice. To bridge this gap, we formulate the data valuation problem in generative models from a similarity-matching perspective. Specifically, we introduce Generative Model Valuator (GMValuator), the first model-agnostic approach for any generative models, designed to provide data valuation for generation tasks. We have conducted extensive experiments to demonstrate the effectiveness of the proposed method. To the best of their knowledge, GMValuator is the first work that offers a training-free, post-hoc data valuation strategy for deep generative models.
♻ ☆ Scalable Set Encoding with Universal Mini-Batch Consistency and Unbiased Full Set Gradient Approximation ICML 2023
Recent work on mini-batch consistency (MBC) for set functions has brought attention to the need for sequentially processing and aggregating chunks of a partitioned set while guaranteeing the same output for all partitions. However, existing constraints on MBC architectures lead to models with limited expressive power. Additionally, prior work has not addressed how to deal with large sets during training when the full set gradient is required. To address these issues, we propose a Universally MBC (UMBC) class of set functions which can be used in conjunction with arbitrary non-MBC components while still satisfying MBC, enabling a wider range of function classes to be used in MBC settings. Furthermore, we propose an efficient MBC training algorithm which gives an unbiased approximation of the full set gradient and has a constant memory overhead for any set size for both train- and test-time. We conduct extensive experiments including image completion, text classification, unsupervised clustering, and cancer detection on high-resolution images to verify the efficiency and efficacy of our scalable set encoding framework.
comment: ICML 2023
♻ ☆ Dynamic Spatial Sparsification for Efficient Vision Transformers and Convolutional Neural Networks NeurIPS 2021
In this paper, we present a new approach for model acceleration by exploiting spatial sparsity in visual data. We observe that the final prediction in vision Transformers is only based on a subset of the most informative tokens, which is sufficient for accurate image recognition. Based on this observation, we propose a dynamic token sparsification framework to prune redundant tokens progressively and dynamically based on the input to accelerate vision Transformers. Specifically, we devise a lightweight prediction module to estimate the importance score of each token given the current features. The module is added to different layers to prune redundant tokens hierarchically. While the framework is inspired by our observation of the sparse attention in vision Transformers, we find the idea of adaptive and asymmetric computation can be a general solution for accelerating various architectures. We extend our method to hierarchical models including CNNs and hierarchical vision Transformers as well as more complex dense prediction tasks that require structured feature maps by formulating a more generic dynamic spatial sparsification framework with progressive sparsification and asymmetric computation for different spatial locations. By applying lightweight fast paths to less informative features and using more expressive slow paths to more important locations, we can maintain the structure of feature maps while significantly reducing the overall computations. Extensive experiments demonstrate the effectiveness of our framework on various modern architectures and different visual recognition tasks. Our results clearly demonstrate that dynamic spatial sparsification offers a new and more effective dimension for model acceleration. Code is available at https://github.com/raoyongming/DynamicViT
comment: Accepted to T-PAMI. Journal version of our NeurIPS 2021 work: arXiv:2106.02034. Code is available at https://github.com/raoyongming/DynamicViT
♻ ☆ Fair Diffusion: Instructing Text-to-Image Generation Models on Fairness
Generative AI models have recently achieved astonishing results in quality and are consequently employed in a fast-growing number of applications. However, since they are highly data-driven, relying on billion-sized datasets randomly scraped from the internet, they also suffer from degenerated and biased human behavior, as we demonstrate. In fact, they may even reinforce such biases. To not only uncover but also combat these undesired effects, we present a novel strategy, called Fair Diffusion, to attenuate biases after the deployment of generative text-to-image models. Specifically, we demonstrate shifting a bias, based on human instructions, in any direction yielding arbitrarily new proportions for, e.g., identity groups. As our empirical evaluation demonstrates, this introduced control enables instructing generative image models on fairness, with no data filtering and additional training required.
♻ ☆ Subject Membership Inference Attacks in Federated Learning
Privacy attacks on Machine Learning (ML) models often focus on inferring the existence of particular data points in the training data. However, what the adversary really wants to know is if a particular individual's (subject's) data was included during training. In such scenarios, the adversary is more likely to have access to the distribution of a particular subject than actual records. Furthermore, in settings like cross-silo Federated Learning (FL), a subject's data can be embodied by multiple data records that are spread across multiple organizations. Nearly all of the existing private FL literature is dedicated to studying privacy at two granularities -- item-level (individual data records), and user-level (participating user in the federation), neither of which apply to data subjects in cross-silo FL. This insight motivates us to shift our attention from the privacy of data records to the privacy of data subjects, also known as subject-level privacy. We propose two novel black-box attacks for subject membership inference, of which one assumes access to a model after each training round. Using these attacks, we estimate subject membership inference risk on real-world data for single-party models as well as FL scenarios. We find our attacks to be extremely potent, even without access to exact training records, and using the knowledge of membership for a handful of subjects. To better understand the various factors that may influence subject privacy risk in cross-silo FL settings, we systematically generate several hundred synthetic federation configurations, varying properties of the data, model design and training, and the federation itself. Finally, we investigate the effectiveness of Differential Privacy in mitigating this threat.
♻ ☆ Supply-Side Equilibria in Recommender Systems
Algorithmic recommender systems such as Spotify and Netflix affect not only consumer behavior but also producer incentives. Producers seek to create content that will be shown by the recommendation algorithm, which can impact both the diversity and quality of their content. In this work, we investigate the resulting supply-side equilibria in personalized content recommender systems. We model users and content as $D$-dimensional vectors, the recommendation algorithm as showing each user the content with highest dot product, and producers as maximizing the number of users who are recommended their content minus the cost of production. Two key features of our model are that the producer decision space is multi-dimensional and the user base is heterogeneous, which contrasts with classical low-dimensional models. Multi-dimensionality and heterogeneity create the potential for specialization, where different producers create different types of content at equilibrium. Using a duality argument, we derive necessary and sufficient conditions for whether specialization occurs: these conditions depend on the extent to which users are heterogeneous and to which producers can perform well on all dimensions at once without incurring a high cost. Then, we characterize the distribution of content at equilibrium in concrete settings with two populations of users. Lastly, we show that specialization can enable producers to achieve positive profit at equilibrium, which means that specialization can reduce the competitiveness of the marketplace. At a conceptual level, our analysis of supply-side competition takes a step towards elucidating how personalized recommendations shape the marketplace of digital goods, and towards understanding what new phenomena arise in multi-dimensional competitive settings.
comment: Updated version with revised and expanded content
♻ ☆ QCBA: Improving Rule Classifiers Learned from Quantitative Data by Recovering Information Lost by Discretisation
A prediscretisation of numerical attributes which is required by some rule learning algorithms is a source of inefficiencies. This paper describes new rule tuning steps that aim to recover lost information in the discretisation and new pruning techniques that may further reduce the size of rule models and improve their accuracy. The proposed QCBA method was initially developed to postprocess quantitative attributes in models generated by the Classification based on associations (CBA) algorithm, but it can also be applied to the results of other rule learning approaches. We demonstrate the effectiveness on the postprocessing of models generated by five association rule classification algorithms (CBA, CMAR, CPAR, IDS, SBRL) and two first-order logic rule learners (FOIL2 and PRM). Benchmarks on 22 datasets from the UCI repository show smaller size and the overall best predictive performance for FOIL2+QCBA compared to all seven baselines. Postoptimised CBA models have a better predictive performance compared to the state-of-the-art rule learner CORELS in this benchmark. The article contains an ablation study for the individual postprocessing steps and a scalability analysis on the KDD'99 Anomaly detection dataset.
comment: online-first. Appl Intell (2023)
♻ ☆ Privacy in Multimodal Federated Human Activity Recognition
Human Activity Recognition (HAR) training data is often privacy-sensitive or held by non-cooperative entities. Federated Learning (FL) addresses such concerns by training ML models on edge clients. This work studies the impact of privacy in federated HAR at a user, environment, and sensor level. We show that the performance of FL for HAR depends on the assumed privacy level of the FL system and primarily upon the colocation of data from different sensors. By avoiding data sharing and assuming privacy at the human or environment level, as prior works have done, the accuracy decreases by 5-7%. However, extending this to the modality level and strictly separating sensor data between multiple clients may decrease the accuracy by 19-42%. As this form of privacy is necessary for the ethical utilisation of passive sensing methods in HAR, we implement a system where clients mutually train both a general FL model and a group-level one per modality. Our evaluation shows that this method leads to only a 7-13% decrease in accuracy, making it possible to build HAR systems with diverse hardware.
comment: In 3rd On-Device Intelligence Workshop at MLSys 2023, 8 pages
♻ ☆ Bayes-optimal limits in structured PCA, and how to reach them
How do statistical dependencies in measurement noise influence high-dimensional inference? To answer this, we study the paradigmatic spiked matrix model of principal components analysis (PCA), where a rank-one matrix is corrupted by additive noise. We go beyond the usual independence assumption on the noise entries, by drawing the noise from a low-order polynomial orthogonal matrix ensemble. The resulting noise correlations make the setting relevant for applications but analytically challenging. We provide the first characterization of the Bayes-optimal limits of inference in this model. If the spike is rotation-invariant, we show that standard spectral PCA is optimal. However, for more general priors, both PCA and the existing approximate message passing algorithm (AMP) fall short of achieving the information-theoretic limits, which we compute using the replica method from statistical mechanics. We thus propose a novel AMP, inspired by the theory of Adaptive Thouless-Anderson-Palmer equations, which saturates the theoretical limit. This AMP comes with a rigorous state evolution analysis tracking its performance. Although we focus on specific noise distributions, our methodology can be generalized to a wide class of trace matrix ensembles at the cost of more involved expressions. Finally, despite the seemingly strong assumption of rotation-invariant noise, our theory empirically predicts algorithmic performance on real data, pointing at remarkable universality properties.
♻ ☆ Fast Nonlinear Vector Quantile Regression
Quantile regression (QR) is a powerful tool for estimating one or more conditional quantiles of a target variable $\mathrm{Y}$ given explanatory features $\boldsymbol{\mathrm{X}}$. A limitation of QR is that it is only defined for scalar target variables, due to the formulation of its objective function, and since the notion of quantiles has no standard definition for multivariate distributions. Recently, vector quantile regression (VQR) was proposed as an extension of QR for vector-valued target variables, thanks to a meaningful generalization of the notion of quantiles to multivariate distributions via optimal transport. Despite its elegance, VQR is arguably not applicable in practice due to several limitations: (i) it assumes a linear model for the quantiles of the target $\boldsymbol{\mathrm{Y}}$ given the features $\boldsymbol{\mathrm{X}}$; (ii) its exact formulation is intractable even for modestly-sized problems in terms of target dimensions, number of regressed quantile levels, or number of features, and its relaxed dual formulation may violate the monotonicity of the estimated quantiles; (iii) no fast or scalable solvers for VQR currently exist. In this work we fully address these limitations, namely: (i) We extend VQR to the non-linear case, showing substantial improvement over linear VQR; (ii) We propose {vector monotone rearrangement}, a method which ensures the quantile functions estimated by VQR are monotone functions; (iii) We provide fast, GPU-accelerated solvers for linear and nonlinear VQR which maintain a fixed memory footprint, and demonstrate that they scale to millions of samples and thousands of quantile levels; (iv) We release an optimized python package of our solvers as to widespread the use of VQR in real-world applications.
comment: 35 pages, 15 figures, code: https://github.com/vistalab-technion/vqr
♻ ☆ Self-Supervised Continual Graph Learning in Adaptive Riemannian Spaces AAAI 2023
Continual graph learning routinely finds its role in a variety of real-world applications where the graph data with different tasks come sequentially. Despite the success of prior works, it still faces great challenges. On the one hand, existing methods work with the zero-curvature Euclidean space, and largely ignore the fact that curvature varies over the coming graph sequence. On the other hand, continual learners in the literature rely on abundant labels, but labeling graph in practice is particularly hard especially for the continuously emerging graphs on-the-fly. To address the aforementioned challenges, we propose to explore a challenging yet practical problem, the self-supervised continual graph learning in adaptive Riemannian spaces. In this paper, we propose a novel self-supervised Riemannian Graph Continual Learner (RieGrace). In RieGrace, we first design an Adaptive Riemannian GCN (AdaRGCN), a unified GCN coupled with a neural curvature adapter, so that Riemannian space is shaped by the learnt curvature adaptive to each graph. Then, we present a Label-free Lorentz Distillation approach, in which we create teacher-student AdaRGCN for the graph sequence. The student successively performs intra-distillation from itself and inter-distillation from the teacher so as to consolidate knowledge without catastrophic forgetting. In particular, we propose a theoretically grounded Generalized Lorentz Projection for the contrastive distillation in Riemannian space. Extensive experiments on the benchmark datasets show the superiority of RieGrace, and additionally, we investigate on how curvature changes over the graph sequence.
comment: Accepted by AAAI 2023 (Main Track), 9 pages, 4 figures
♻ ☆ When Federated Learning Meets Pre-trained Language Models' Parameter-Efficient Tuning Methods
With increasing privacy concerns on data, recent studies have made significant progress using federated learning (FL) on privacy-sensitive natural language processing (NLP) tasks. Much literature suggests fully fine-tuning pre-trained language models (PLMs) in the FL paradigm can mitigate the data heterogeneity problem and close the performance gap with centralized training. However, large PLMs bring the curse of prohibitive communication overhead and local model adaptation costs for the FL system. To this end, we introduce various parameter-efficient tuning (PETuning) methods into federated learning. Specifically, we provide a holistic empirical study of representative PLMs tuning methods in FL. The experimental results cover the analysis of data heterogeneity levels, data scales, and different FL scenarios. Overall communication overhead can be significantly reduced by locally tuning and globally aggregating lightweight model parameters while maintaining acceptable performance in various FL settings. To facilitate the research of PETuning in FL, we also develop a federated tuning framework FedPETuning, which allows practitioners to exploit different PETuning methods under the FL training paradigm conveniently. The source code is available at \url{https://github.com/iezhuozhuo/FedETuning/tree/deltaTuning}.
♻ ☆ Analysis of Perceived Stress Test using Machine Learning
The aim of this study is to determine the perceived stress levels of 150 individuals and analyze the responses given to adapted questions in Turkish using machine learning. The test consists of 14 questions, each scored on a scale of 0 to 4, resulting in a total score range of 0-56. Out of these questions, 7 are formulated in a negative context and scored accordingly, while the remaining 7 are formulated in a positive context and scored in reverse. The test is also designed to identify two sub-factors: perceived self-efficacy and stress/discomfort perception. The main objectives of this research are to demonstrate that test questions may not have equal importance using artificial intelligence techniques, reveal which questions exhibit variations in the society using machine learning, and ultimately demonstrate the existence of distinct patterns observed psychologically. This study provides a different perspective from the existing psychology literature by repeating the test through machine learning. Additionally, it questions the accuracy of the scale used to interpret the results of the perceived stress test and emphasizes the importance of considering differences in the prioritization of test questions. The findings of this study offer new insights into coping strategies and therapeutic approaches in dealing with stress. Source code: https://github.com/toygarr/ppl-r-stressed
comment: in Turkish language
♻ ☆ Margin-based sampling in high dimensions: When being active is less efficient than staying passive
It is widely believed that given the same labeling budget, active learning (AL) algorithms like margin-based active learning achieve better predictive performance than passive learning (PL), albeit at a higher computational cost. Recent empirical evidence suggests that this added cost might be in vain, as margin-based AL can sometimes perform even worse than PL. While existing works offer different explanations in the low-dimensional regime, this paper shows that the underlying mechanism is entirely different in high dimensions: we prove for logistic regression that PL outperforms margin-based AL even for noiseless data and when using the Bayes optimal decision boundary for sampling. Insights from our proof indicate that this high-dimensional phenomenon is exacerbated when the separation between the classes is small. We corroborate this intuition with experiments on 20 high-dimensional datasets spanning a diverse range of applications, from finance and histology to chemistry and computer vision.
♻ ☆ Hindsight Learning for MDPs with Exogenous Inputs
Many resource management problems require sequential decision-making under uncertainty, where the only uncertainty affecting the decision outcomes are exogenous variables outside the control of the decision-maker. We model these problems as Exo-MDPs (Markov Decision Processes with Exogenous Inputs) and design a class of data-efficient algorithms for them termed Hindsight Learning (HL). Our HL algorithms achieve data efficiency by leveraging a key insight: having samples of the exogenous variables, past decisions can be revisited in hindsight to infer counterfactual consequences that can accelerate policy improvements. We compare HL against classic baselines in the multi-secretary and airline revenue management problems. We also scale our algorithms to a business-critical cloud resource management problem -- allocating Virtual Machines (VMs) to physical machines, and simulate their performance with real datasets from a large public cloud provider. We find that HL algorithms outperform domain-specific heuristics, as well as state-of-the-art reinforcement learning methods.
comment: 53 pages, 6 figures
♻ ☆ ChatGPT: Jack of all trades, master of none
OpenAI has released the Chat Generative Pre-trained Transformer (ChatGPT) and revolutionized the approach in artificial intelligence to human-model interaction. Several publications on ChatGPT evaluation test its effectiveness on well-known natural language processing (NLP) tasks. However, the existing studies are mostly non-automated and tested on a very limited scale. In this work, we examined ChatGPT's capabilities on 25 diverse analytical NLP tasks, most of them subjective even to humans, such as sentiment analysis, emotion recognition, offensiveness, and stance detection. In contrast, the other tasks require more objective reasoning like word sense disambiguation, linguistic acceptability, and question answering. We also evaluated GPT-4 model on five selected subsets of NLP tasks. We automated ChatGPT and GPT-4 prompting process and analyzed more than 49k responses. Our comparison of its results with available State-of-the-Art (SOTA) solutions showed that the average loss in quality of the ChatGPT model was about 25% for zero-shot and few-shot evaluation. For GPT-4 model, a loss for semantic tasks is significantly lower than for ChatGPT. We showed that the more difficult the task (lower SOTA performance), the higher the ChatGPT loss. It especially refers to pragmatic NLP problems like emotion recognition. We also tested the ability to personalize ChatGPT responses for selected subjective tasks via Random Contextual Few-Shot Personalization, and we obtained significantly better user-based predictions. Additional qualitative analysis revealed a ChatGPT bias, most likely due to the rules imposed on human trainers by OpenAI. Our results provide the basis for a fundamental discussion of whether the high quality of recent predictive NLP models can indicate a tool's usefulness to society and how the learning and validation procedures for such systems should be established.
comment: preprint
♻ ☆ Truncated Affinity Maximization: One-class Homophily Modeling for Graph Anomaly Detection
One prevalent property we find empirically in real-world graph anomaly detection (GAD) datasets is a one-class homophily, i.e., normal nodes tend to have strong connection/affinity with each other, while the homophily in abnormal nodes is significantly weaker than normal nodes. However, this anomaly-discriminative property is ignored by existing GAD methods that are typically built using a conventional anomaly detection objective, such as data reconstruction. In this work, we explore this property to introduce a novel unsupervised anomaly scoring measure for GAD -- local node affinity -- that assigns a larger anomaly score to nodes that are less affiliated with their neighbors, with the affinity defined as similarity on node attributes/representations. We further propose Truncated Affinity Maximization (TAM) that learns tailored node representations for our anomaly measure by maximizing the local affinity of nodes to their neighbors. Optimizing on the original graph structure can be biased by non-homophily edges (i.e., edges connecting normal and abnormal nodes). Thus, TAM is instead optimized on truncated graphs where non-homophily edges are removed iteratively to mitigate this bias. The learned representations result in significantly stronger local affinity for normal nodes than abnormal nodes. Extensive empirical results on six real-world GAD datasets show that TAM substantially outperforms seven competing models, achieving over 10% increase in AUROC/AUPRC compared to the best contenders on challenging datasets. Our code will be made available at https: //github.com/mala-lab/TAM-master/.
comment: 19 pages, 9 figures
♻ ☆ Reward is enough for convex MDPs
Maximising a cumulative reward function that is Markov and stationary, i.e., defined over state-action pairs and independent of time, is sufficient to capture many kinds of goals in a Markov decision process (MDP). However, not all goals can be captured in this manner. In this paper we study convex MDPs in which goals are expressed as convex functions of the stationary distribution and show that they cannot be formulated using stationary reward functions. Convex MDPs generalize the standard reinforcement learning (RL) problem formulation to a larger framework that includes many supervised and unsupervised RL problems, such as apprenticeship learning, constrained MDPs, and so-called `pure exploration'. Our approach is to reformulate the convex MDP problem as a min-max game involving policy and cost (negative reward) `players', using Fenchel duality. We propose a meta-algorithm for solving this problem and show that it unifies many existing algorithms in the literature.
♻ ☆ Smooth-Trajectron++: Augmenting the Trajectron++ behaviour prediction model with smooth attention
Understanding traffic participants' behaviour is crucial for predicting their future trajectories, aiding in developing safe and reliable planning systems for autonomous vehicles. Integrating cognitive processes and machine learning models has shown promise in other domains but is lacking in the trajectory forecasting of multiple traffic agents in large-scale autonomous driving datasets. This work investigates the state-of-the-art trajectory forecasting model Trajectron++ which we enhance by incorporating a smoothing term in its attention module. This attention mechanism mimics human attention inspired by cognitive science research indicating limits to attention switching. We evaluate the performance of the resulting Smooth-Trajectron++ model and compare it to the original model on various benchmarks, revealing the potential of incorporating insights from human cognition into trajectory prediction models.
♻ ☆ Efficient Algorithms for Exact Graph Matching on Correlated Stochastic Block Models with Constant Correlation ICML 2023
We consider the problem of graph matching, or learning vertex correspondence, between two correlated stochastic block models (SBMs). The graph matching problem arises in various fields, including computer vision, natural language processing and bioinformatics, and in particular, matching graphs with inherent community structure has significance related to de-anonymization of correlated social networks. Compared to the correlated Erdos-Renyi (ER) model, where various efficient algorithms have been developed, among which a few algorithms have been proven to achieve the exact matching with constant edge correlation, no low-order polynomial algorithm has been known to achieve exact matching for the correlated SBMs with constant correlation. In this work, we propose an efficient algorithm for matching graphs with community structure, based on the comparison between partition trees rooted from each vertex, by extending the idea of Mao et al. (2021) to graphs with communities. The partition tree divides the large neighborhoods of each vertex into disjoint subsets using their edge statistics to different communities. Our algorithm is the first low-order polynomial-time algorithm achieving exact matching between two correlated SBMs with high probability in dense graphs.
comment: ICML 2023
♻ ☆ One Risk to Rule Them All: Addressing Distributional Shift in Offline Reinforcement Learning via Risk-Aversion
Offline reinforcement learning (RL) is suitable for safety-critical domains where online exploration is not feasible. In such domains, decision-making should take into consideration the risk of catastrophic outcomes. In other words, decision-making should be risk-averse. An additional challenge of offline RL is avoiding distributional shift, i.e. ensuring that state-action pairs visited by the policy remain near those in the dataset. Previous works on risk in offline RL combine offline RL techniques (to avoid distributional shift), with risk-sensitive RL algorithms (to achieve risk-aversion). In this work, we propose risk-aversion as a mechanism to jointly address both of these issues. We propose a model-based approach, and use an ensemble of models to estimate epistemic uncertainty, in addition to aleatoric uncertainty. We train a policy that is risk-averse, and avoids high uncertainty actions. Risk-aversion to epistemic uncertainty prevents distributional shift, as areas not covered by the dataset have high epistemic uncertainty. Risk-aversion to aleatoric uncertainty discourages actions that are inherently risky due to environment stochasticity. Thus, by only introducing risk-aversion, we avoid distributional shift in addition to achieving risk-aversion to aleatoric risk. Our algorithm, 1R2R, achieves strong performance on deterministic benchmarks, and outperforms existing approaches for risk-sensitive objectives in stochastic domains.
♻ ☆ Learning Physically Realizable Skills for Online Packing of General 3D Shapes
We study the problem of learning online packing skills for irregular 3D shapes, which is arguably the most challenging setting of bin packing problems. The goal is to consecutively move a sequence of 3D objects with arbitrary shapes into a designated container with only partial observations of the object sequence. Meanwhile, we take physical realizability into account, involving physics dynamics and constraints of a placement. The packing policy should understand the 3D geometry of the object to be packed and make effective decisions to accommodate it in the container in a physically realizable way. We propose a Reinforcement Learning (RL) pipeline to learn the policy. The complex irregular geometry and imperfect object placement together lead to huge solution space. Direct training in such space is prohibitively data intensive. We instead propose a theoretically-provable method for candidate action generation to reduce the action space of RL and the learning burden. A parameterized policy is then learned to select the best placement from the candidates. Equipped with an efficient method of asynchronous RL acceleration and a data preparation process of simulation-ready training sequences, a mature packing policy can be trained in a physics-based environment within 48 hours. Through extensive evaluation on a variety of real-life shape datasets and comparisons with state-of-the-art baselines, we demonstrate that our method outperforms the best-performing baseline on all datasets by at least 12.8% in terms of packing utility.
comment: ACM Transactions on Graphics (TOG)
♻ ☆ Differentiable Multi-Target Causal Bayesian Experimental Design ICML 2023
We introduce a gradient-based approach for the problem of Bayesian optimal experimental design to learn causal models in a batch setting -- a critical component for causal discovery from finite data where interventions can be costly or risky. Existing methods rely on greedy approximations to construct a batch of experiments while using black-box methods to optimize over a single target-state pair to intervene with. In this work, we completely dispose of the black-box optimization techniques and greedy heuristics and instead propose a conceptually simple end-to-end gradient-based optimization procedure to acquire a set of optimal intervention target-state pairs. Such a procedure enables parameterization of the design space to efficiently optimize over a batch of multi-target-state interventions, a setting which has hitherto not been explored due to its complexity. We demonstrate that our proposed method outperforms baselines and existing acquisition strategies in both single-target and multi-target settings across a number of synthetic datasets.
comment: Camera-ready version ICML 2023
♻ ☆ BotArtist: Twitter bot detection Machine Learning model based on Twitter suspension
Twitter as one of the most popular social networks, offers a means for communication and online discourse, which unfortunately has been the target of bots and fake accounts, leading to the manipulation and spreading of false information. Towards this end, we gather a challenging, multilingual dataset of social discourse on Twitter, originating from 9M users regarding the recent Russo-Ukrainian war, in order to detect the bot accounts and the conversation involving them. We collect the ground truth for our dataset through the Twitter API suspended accounts collection, containing approximately 343K of bot accounts and 8M of normal users. Additionally, we use a dataset provided by Botometer-V3 with 1,777 Varol, 483 German accounts, and 1,321 US accounts. Besides the publicly available datasets, we also manage to collect 2 independent datasets around popular discussion topics of the 2022 energy crisis and the 2022 conspiracy discussions. Both of the datasets were labeled according to the Twitter suspension mechanism. We build a novel ML model for bot detection using the state-of-the-art XGBoost model. We combine the model with a high volume of labeled tweets according to the Twitter suspension mechanism ground truth. This requires a limited set of profile features allowing labeling of the dataset in different time periods from the collection, as it is independent of the Twitter API. In comparison with Botometer our methodology achieves an average 11% higher ROC-AUC score over two real-case scenario datasets.
♻ ☆ Lessons on Parameter Sharing across Layers in Transformers
We propose a parameter sharing method for Transformers (Vaswani et al., 2017). The proposed approach relaxes a widely used technique, which shares parameters for one layer with all layers such as Universal Transformers (Dehghani et al., 2019), to increase the efficiency in the computational time. We propose three strategies: Sequence, Cycle, and Cycle (rev) to assign parameters to each layer. Experimental results show that the proposed strategies are efficient in the parameter size and computational time. Moreover, we indicate that the proposed strategies are also effective in the configuration where we use many training data such as the recent WMT competition.
comment: SustaiNLP 2023
♻ ☆ Learning Practically Feasible Policies for Online 3D Bin Packing
We tackle the Online 3D Bin Packing Problem, a challenging yet practically useful variant of the classical Bin Packing Problem. In this problem, the items are delivered to the agent without informing the full sequence information. Agent must directly pack these items into the target bin stably without changing their arrival order, and no further adjustment is permitted. Online 3D-BPP can be naturally formulated as Markov Decision Process (MDP). We adopt deep reinforcement learning, in particular, the on-policy actor-critic framework, to solve this MDP with constrained action space. To learn a practically feasible packing policy, we propose three critical designs. First, we propose an online analysis of packing stability based on a novel stacking tree. It attains a high analysis accuracy while reducing the computational complexity from $O(N^2)$ to $O(N \log N)$, making it especially suited for RL training. Second, we propose a decoupled packing policy learning for different dimensions of placement which enables high-resolution spatial discretization and hence high packing precision. Third, we introduce a reward function that dictates the robot to place items in a far-to-near order and therefore simplifies the collision avoidance in movement planning of the robotic arm. Furthermore, we provide a comprehensive discussion on several key implemental issues. The extensive evaluation demonstrates that our learned policy outperforms the state-of-the-art methods significantly and is practically usable for real-world applications.
comment: Science China Information Sciences
♻ ☆ Getting ViT in Shape: Scaling Laws for Compute-Optimal Model Design
Scaling laws have been recently employed to derive compute-optimal model size (number of parameters) for a given compute duration. We advance and refine such methods to infer compute-optimal model shapes, such as width and depth, and successfully implement this in vision transformers. Our shape-optimized vision transformer, SoViT, achieves results competitive with models that exceed twice its size, despite being pre-trained with an equivalent amount of compute. For example, SoViT-400m/14 achieves 90.3% fine-tuning accuracy on ILSRCV2012, surpassing the much larger ViT-g/14 and approaching ViT-G/14 under identical settings, with also less than half the inference cost. We conduct a thorough evaluation across multiple tasks, such as image classification, captioning, VQA and zero-shot transfer, demonstrating the effectiveness of our model across a broad range of domains and identifying limitations. Overall, our findings challenge the prevailing approach of blindly scaling up vision models and pave a path for a more informed scaling.
comment: 10 pages, 7 figures, 9 tables. Version 2: Layout fixes
♻ ☆ Do intermediate feature coalitions aid explainability of black-box models?
This work introduces the notion of intermediate concepts based on levels structure to aid explainability for black-box models. The levels structure is a hierarchical structure in which each level corresponds to features of a dataset (i.e., a player-set partition). The level of coarseness increases from the trivial set, which only comprises singletons, to the set, which only contains the grand coalition. In addition, it is possible to establish meronomies, i.e., part-whole relationships, via a domain expert that can be utilised to generate explanations at an abstract level. We illustrate the usability of this approach in a real-world car model example and the Titanic dataset, where intermediate concepts aid in explainability at different levels of abstraction.
comment: 14 pages,The 1st World Conference on eXplainable Artificial Intelligence, 2023
♻ ☆ ByzSecAgg: A Byzantine-Resistant Secure Aggregation Scheme for Federated Learning Based on Coded Computing and Vector Commitment
In this paper, we propose ByzSecAgg, an efficient secure aggregation scheme for federated learning that is protected against Byzantine attacks and privacy leakages. Processing individual updates to manage adversarial behavior, while preserving privacy of data against colluding nodes, requires some sort of secure secret sharing. However, the communication load for secret sharing of long vectors of updates can be very high. ByzSecAgg solves this problem by partitioning local updates into smaller sub-vectors and sharing them using ramp secret sharing. However, this sharing method does not admit bi-linear computations, such as pairwise distance calculations, needed by outlier-detection algorithms. To overcome this issue, each user runs another round of ramp sharing, with different embedding of data in the sharing polynomial. This technique, motivated by ideas from coded computing, enables secure computation of pairwise distance. In addition, to maintain the integrity and privacy of the local update, ByzSecAgg also uses a vector commitment method, in which the commitment size remains constant (i.e. does not increase with the length of the local update), while simultaneously allowing verification of the secret sharing process. In terms of communication loads, ByzSecAgg significantly outperforms the state-of-the-art scheme, known as BREA.
♻ ☆ Generalization with Reverse-Calibration of Well and Seismic Data Using Machine Learning Methods for Complex Reservoirs Predicting During Early-Stage Geological Exploration Oil Field
The aim of this study is to develop and apply an autonomous approach for predicting the probability of hydrocarbon reservoirs spreading in the studied area. The methodology uses machine learning algorithms in the problem of binary classification, which restore the probability function of the space element belonging to the classes identified by the results of interpretation of well logging. Attributes of seismic wavefield are used as predictors. The study includes the following sequence of actions: creation of data sets for training, selection of features, reverse-calibration of data, creation of a population of classification models, evaluation of classification quality, evaluation of the contribution of features in the prediction, ensembling the population of models by stacking method. As a result, a prediction was made - a three-dimensional cube of calibrated probabilities of belonging of the studied space to the class of reservoir and its derivative in the form of the map of reservoir thicknesses of the Achimov complex of deposits was obtained. Assessment of changes in the quality of the forecast depending on the use of different data sets was carried out. Conclusion. The reverse-calibration method proposed in this work uses the uncertainty of geophysical data as a hyperparameter of the global tuning of the technological stack, within the given limits of the a priori error of these data. It is shown that the method improves the quality of the forecast. The technological stack of machine learning algorithms used in this work allows expert-independent generalization of geological and geophysical data, and use this generalization to test hypotheses and create geological models based on a probabilistic view of the reservoir.
comment: 12 pages, 1 figure
♻ ☆ Probably Anytime-Safe Stochastic Combinatorial Semi-Bandits ICML 2023
Motivated by concerns about making online decisions that incur undue amount of risk at each time step, in this paper, we formulate the probably anytime-safe stochastic combinatorial semi-bandits problem. In this problem, the agent is given the option to select a subset of size at most $K$ from a set of $L$ ground items. Each item is associated to a certain mean reward as well as a variance that represents its risk. To mitigate the risk that the agent incurs, we require that with probability at least $1-\delta$, over the entire horizon of time $T$, each of the choices that the agent makes should contain items whose sum of variances does not exceed a certain variance budget. We call this probably anytime-safe constraint. Under this constraint, we design and analyze an algorithm {\sc PASCombUCB} that minimizes the regret over the horizon of time $T$. By developing accompanying information-theoretic lower bounds, we show that under both the problem-dependent and problem-independent paradigms, {\sc PASCombUCB} is almost asymptotically optimal. Experiments are conducted to corroborate our theoretical findings. Our problem setup, the proposed {\sc PASCombUCB} algorithm, and novel analyses are applicable to domains such as recommendation systems and transportation in which an agent is allowed to choose multiple items at a single time step and wishes to control the risk over the whole time horizon.
comment: To be presented at ICML 2023. 57 pages, 6 figures
♻ ☆ Learning from Physical Human Feedback: An Object-Centric One-Shot Adaptation Method ICRA 2023
For robots to be effectively deployed in novel environments and tasks, they must be able to understand the feedback expressed by humans during intervention. This can either correct undesirable behavior or indicate additional preferences. Existing methods either require repeated episodes of interactions or assume prior known reward features, which is data-inefficient and can hardly transfer to new tasks. We relax these assumptions by describing human tasks in terms of object-centric sub-tasks and interpreting physical interventions in relation to specific objects. Our method, Object Preference Adaptation (OPA), is composed of two key stages: 1) pre-training a base policy to produce a wide variety of behaviors, and 2) online-updating according to human feedback. The key to our fast, yet simple adaptation is that general interaction dynamics between agents and objects are fixed, and only object-specific preferences are updated. Our adaptation occurs online, requires only one human intervention (one-shot), and produces new behaviors never seen during training. Trained on cheap synthetic data instead of expensive human demonstrations, our policy correctly adapts to human perturbations on realistic tasks on a physical 7DOF robot. Videos, code, and supplementary material are provided.
comment: Accepted to ICRA 2023
♻ ☆ Unbiased constrained sampling with Self-Concordant Barrier Hamiltonian Monte Carlo
In this paper, we propose Barrier Hamiltonian Monte Carlo (BHMC), a version of the HMC algorithm which aims at sampling from a Gibbs distribution $\pi$ on a manifold $\mathrm{M}$, endowed with a Hessian metric $\mathfrak{g}$ derived from a self-concordant barrier. Our method relies on Hamiltonian dynamics which comprises $\mathfrak{g}$. Therefore, it incorporates the constraints defining $\mathrm{M}$ and is able to exploit its underlying geometry. However, the corresponding Hamiltonian dynamics is defined via non separable Ordinary Differential Equations (ODEs) in contrast to the Euclidean case. It implies unavoidable bias in existing generalization of HMC to Riemannian manifolds. In this paper, we propose a new filter step, called "involution checking step", to address this problem. This step is implemented in two versions of BHMC, coined continuous BHMC (c-BHMC) and numerical BHMC (n-BHMC) respectively. Our main results establish that these two new algorithms generate reversible Markov chains with respect to $\pi$ and do not suffer from any bias in comparison to previous implementations. Our conclusions are supported by numerical experiments where we consider target distributions defined on polytopes.
♻ ☆ Differentiable and Transportable Structure Learning ICML
Directed acyclic graphs (DAGs) encode a lot of information about a particular distribution in their structure. However, compute required to infer these structures is typically super-exponential in the number of variables, as inference requires a sweep of a combinatorially large space of potential structures. That is, until recent advances made it possible to search this space using a differentiable metric, drastically reducing search time. While this technique -- named NOTEARS -- is widely considered a seminal work in DAG-discovery, it concedes an important property in favour of differentiability: transportability. To be transportable, the structures discovered on one dataset must apply to another dataset from the same domain. We introduce D-Struct which recovers transportability in the discovered structures through a novel architecture and loss function while remaining fully differentiable. Because D-Struct remains differentiable, our method can be easily adopted in existing differentiable architectures, as was previously done with NOTEARS. In our experiments, we empirically validate D-Struct with respect to edge accuracy and structural Hamming distance in a variety of settings.
comment: Accepted at the International Conference on Machine Learning (ICML) 2023
♻ ☆ Local Model Reconstruction Attacks in Federated Learning and their Uses
In this paper, we initiate the study of local model reconstruction attacks for federated learning, where a honest-but-curious adversary eavesdrops the messages exchanged between a targeted client and the server, and then reconstructs the local/personalized model of the victim. The local model reconstruction attack allows the adversary to trigger other classical attacks in a more effective way, since the local model only depends on the client's data and can leak more private information than the global model learned by the server. Additionally, we propose a novel model-based attribute inference attack in federated learning leveraging the local model reconstruction attack. We provide an analytical lower-bound for this attribute inference attack. Empirical results using real world datasets confirm that our local reconstruction attack works well for both regression and classification tasks. Moreover, we benchmark our novel attribute inference attack against the state-of-the-art attacks in federated learning. Our attack results in higher reconstruction accuracy especially when the clients' datasets are heterogeneous. Our work provides a new angle for designing powerful and explainable attacks to effectively quantify the privacy risk in FL.
♻ ☆ Joint Representations for Reinforcement Learning with Multiple Sensors
Combining inputs from multiple sensor modalities effectively in reinforcement learning (RL) is an open problem. While many self-supervised representation learning approaches exist to improve performance and sample complexity for image-based RL, they usually neglect other available information, such as robot proprioception. However, using this proprioception for representation learning can help algorithms to focus on relevant aspects and guide them toward finding better representations. In this work, we systematically analyze representation learning for RL from multiple sensors by building on Recurrent State Space Models. We propose a combination of reconstruction-based and contrastive losses, which allows us to choose the most appropriate method for each sensor modality. We demonstrate the benefits of joint representations, particularly with distinct loss functions for each modality, for model-free and model-based RL on complex tasks. Those include tasks where the images contain distractions or occlusions and a new locomotion suite. We show that combining reconstruction-based and contrastive losses for joint representation learning improves performance significantly compared to a post hoc combination of image representations and proprioception and can also improve the quality of learned models for model-based RL.
♻ ☆ The Lost Art of Mathematical Modelling
We provide a critique of mathematical biology in light of rapid developments in modern machine learning. We argue that out of the three modelling activities -- (1) formulating models; (2) analysing models; and (3) fitting or comparing models to data -- inherent to mathematical biology, researchers currently focus too much on activity (2) at the cost of (1). This trend, we propose, can be reversed by realising that any given biological phenomena can be modelled in an infinite number of different ways, through the adoption of an open/pluralistic approach. We explain the open approach using fish locomotion as a case study and illustrate some of the pitfalls -- universalism, creating models of models, etc. -- that hinder mathematical biology. We then ask how we might rediscover a lost art: that of creative mathematical modelling. This article is dedicated to the memory of Edmund Crampin.
♻ ☆ LiT-4-RSVQA: Lightweight Transformer-based Visual Question Answering in Remote Sensing
Visual question answering (VQA) methods in remote sensing (RS) aim to answer natural language questions with respect to an RS image. Most of the existing methods require a large amount of computational resources, which limits their application in operational scenarios in RS. To address this issue, in this paper we present an effective lightweight transformer-based VQA in RS (LiT-4-RSVQA) architecture for efficient and accurate VQA in RS. Our architecture consists of: i) a lightweight text encoder module; ii) a lightweight image encoder module; iii) a fusion module; and iv) a classification module. The experimental results obtained on a VQA benchmark dataset demonstrate that our proposed LiT-4-RSVQA architecture provides accurate VQA results while significantly reducing the computational requirements on the executing hardware. Our code is publicly available at https://git.tu-berlin.de/rsim/lit4rsvqa.
comment: Accepted at IEEE International Geoscience and Remote Sensing Symposium 2023
♻ ☆ Decentralized Online Regularized Learning Over Random Time-Varying Graphs
We study the decentralized online regularized linear regression algorithm over random time-varying graphs. At each time step, every node runs an online estimation algorithm consisting of an innovation term processing its own new measurement, a consensus term taking a weighted sum of estimations of its own and its neighbors with additive and multiplicative communication noises and a regularization term preventing over-fitting. It is not required that the regression matrices and graphs satisfy special statistical assumptions such as mutual independence, spatio-temporal independence or stationarity. We develop the nonnegative supermartingale inequality of the estimation error, and prove that the estimations of all nodes converge to the unknown true parameter vector almost surely if the algorithm gains, graphs and regression matrices jointly satisfy the sample path spatio-temporal persistence of excitation condition. Especially, this condition holds by choosing appropriate algorithm gains if the graphs are uniformly conditionally jointly connected and conditionally balanced, and the regression models of all nodes are uniformly conditionally spatio-temporally jointly observable, under which the algorithm converges in mean square and almost surely. In addition, we prove that the regret upper bound is $O(T^{1-\tau}\ln T)$, where $\tau\in (0.5,1)$ is a constant depending on the algorithm gains.
♻ ☆ Differentiable Programming for Earth System Modeling
Earth System Models (ESMs) are the primary tools for investigating future Earth system states at time scales from decades to centuries, especially in response to anthropogenic greenhouse gas release. State-of-the-art ESMs can reproduce the observational global mean temperature anomalies of the last 150 years. Nevertheless, ESMs need further improvements, most importantly regarding (i) the large spread in their estimates of climate sensitivity, i.e., the temperature response to increases in atmospheric greenhouse gases, (ii) the modeled spatial patterns of key variables such as temperature and precipitation, (iii) their representation of extreme weather events, and (iv) their representation of multistable Earth system components and their ability to predict associated abrupt transitions. Here, we argue that making ESMs automatically differentiable has huge potential to advance ESMs, especially with respect to these key shortcomings. First, automatic differentiability would allow objective calibration of ESMs, i.e., the selection of optimal values with respect to a cost function for a large number of free parameters, which are currently tuned mostly manually. Second, recent advances in Machine Learning (ML) and in the amount, accuracy, and resolution of observational data promise to be helpful with at least some of the above aspects because ML may be used to incorporate additional information from observations into ESMs. Automatic differentiability is an essential ingredient in the construction of such hybrid models, combining process-based ESMs with ML components. We document recent work showcasing the potential of automatic differentiation for a new generation of substantially improved, data-informed ESMs.
comment: 17 pages, 2 figures
♻ ☆ Your Contrastive Learning Is Secretly Doing Stochastic Neighbor Embedding ICLR 2023
Contrastive learning, especially self-supervised contrastive learning (SSCL), has achieved great success in extracting powerful features from unlabeled data. In this work, we contribute to the theoretical understanding of SSCL and uncover its connection to the classic data visualization method, stochastic neighbor embedding (SNE), whose goal is to preserve pairwise distances. From the perspective of preserving neighboring information, SSCL can be viewed as a special case of SNE with the input space pairwise similarities specified by data augmentation. The established correspondence facilitates deeper theoretical understanding of learned features of SSCL, as well as methodological guidelines for practical improvement. Specifically, through the lens of SNE, we provide novel analysis on domain-agnostic augmentations, implicit bias and robustness of learned features. To illustrate the practical advantage, we demonstrate that the modifications from SNE to $t$-SNE can also be adopted in the SSCL setting, achieving significant improvement in both in-distribution and out-of-distribution generalization.
comment: Accepted by ICLR 2023
♻ ☆ On the Effectiveness of Hybrid Mutual Information Estimation
Estimating the mutual information from samples from a joint distribution is a challenging problem in both science and engineering. In this work, we realize a variational bound that generalizes both discriminative and generative approaches. Using this bound, we propose a hybrid method to mitigate their respective shortcomings. Further, we propose Predictive Quantization (PQ): a simple generative method that can be easily combined with discriminative estimators for minimal computational overhead. Our propositions yield a tighter bound on the information thanks to the reduced variance of the estimator. We test our methods on a challenging task of correlated high-dimensional Gaussian distributions and a stochastic process involving a system of free particles subjected to a fixed energy landscape. Empirical results show that hybrid methods consistently improved mutual information estimates when compared to the corresponding discriminative counterpart.
♻ ☆ DiffPhase: Generative Diffusion-based STFT Phase Retrieval ICASSP 2023
Diffusion probabilistic models have been recently used in a variety of tasks, including speech enhancement and synthesis. As a generative approach, diffusion models have been shown to be especially suitable for imputation problems, where missing data is generated based on existing data. Phase retrieval is inherently an imputation problem, where phase information has to be generated based on the given magnitude. In this work we build upon previous work in the speech domain, adapting a speech enhancement diffusion model specifically for STFT phase retrieval. Evaluation using speech quality and intelligibility metrics shows the diffusion approach is well-suited to the phase retrieval task, with performance surpassing both classical and modern methods.
comment: Accepted by ICASSP 2023
♻ ☆ De Rham compatible Deep Neural Network FEM
On general regular simplicial partitions $\mathcal{T}$ of bounded polytopal domains $\Omega \subset \mathbb{R}^d$, $d\in\{2,3\}$, we construct \emph{exact neural network (NN) emulations} of all lowest order finite element spaces in the discrete de Rham complex. These include the spaces of piecewise constant functions, continuous piecewise linear (CPwL) functions, the classical ``Raviart-Thomas element'', and the ``N\'{e}d\'{e}lec edge element''. For all but the CPwL case, our network architectures employ both ReLU (rectified linear unit) and BiSU (binary step unit) activations to capture discontinuities. In the important case of CPwL functions, we prove that it suffices to work with pure ReLU nets. Our construction and DNN architecture generalizes previous results in that no geometric restrictions on the regular simplicial partitions $\mathcal{T}$ of $\Omega$ are required for DNN emulation. In addition, for CPwL functions our DNN construction is valid in any dimension $d\geq 2$. Our ``FE-Nets'' are required in the variationally correct, structure-preserving approximation of boundary value problems of electromagnetism in nonconvex polyhedra $\Omega \subset \mathbb{R}^3$. They are thus an essential ingredient in the application of e.g., the methodology of ``physics-informed NNs'' or ``deep Ritz methods'' to electromagnetic field simulation via deep learning techniques. We indicate generalizations of our constructions to higher-order compatible spaces and other, non-compatible classes of discretizations, in particular the ``Crouzeix-Raviart'' elements and Hybridized, Higher Order (HHO) methods.
♻ ☆ SWL-Adapt: An Unsupervised Domain Adaptation Model with Sample Weight Learning for Cross-User Wearable Human Activity Recognition AAAI 2023
In practice, Wearable Human Activity Recognition (WHAR) models usually face performance degradation on the new user due to user variance. Unsupervised domain adaptation (UDA) becomes the natural solution to cross-user WHAR under annotation scarcity. Existing UDA models usually align samples across domains without differentiation, which ignores the difference among samples. In this paper, we propose an unsupervised domain adaptation model with sample weight learning (SWL-Adapt) for cross-user WHAR. SWL-Adapt calculates sample weights according to the classification loss and domain discrimination loss of each sample with a parameterized network. We introduce the meta-optimization based update rule to learn this network end-to-end, which is guided by meta-classification loss on the selected pseudo-labeled target samples. Therefore, this network can fit a weighting function according to the cross-user WHAR task at hand, which is superior to existing sample differentiation rules fixed for special scenarios. Extensive experiments on three public WHAR datasets demonstrate that SWL-Adapt achieves the state-of-the-art performance on the cross-user WHAR task, outperforming the best baseline by an average of 3.1% and 5.3% in accuracy and macro F1 score, respectively.
comment: Accepted by AAAI 2023. 9 pages and 4 figures in main text. 3 pages and 5 figures in appendix
♻ ☆ Towards Understanding Generalization of Macro-AUC in Multi-label Learning ICML 2023
Macro-AUC is the arithmetic mean of the class-wise AUCs in multi-label learning and is commonly used in practice. However, its theoretical understanding is far lacking. Toward solving it, we characterize the generalization properties of various learning algorithms based on the corresponding surrogate losses w.r.t. Macro-AUC. We theoretically identify a critical factor of the dataset affecting the generalization bounds: \emph{the label-wise class imbalance}. Our results on the imbalance-aware error bounds show that the widely-used univariate loss-based algorithm is more sensitive to the label-wise class imbalance than the proposed pairwise and reweighted loss-based ones, which probably implies its worse performance. Moreover, empirical results on various datasets corroborate our theory findings. To establish it, technically, we propose a new (and more general) McDiarmid-type concentration inequality, which may be of independent interest.
comment: ICML 2023
♻ ☆ Is Distance Matrix Enough for Geometric Deep Learning?
Graph Neural Networks (GNNs) are often used for tasks involving the 3D geometry of a given graph, such as molecular dynamics simulation. Although the distance matrix of a geometric graph contains complete geometric information, it has been demonstrated that Message Passing Neural Networks (MPNNs) are insufficient for learning this geometry. In this work, we expand on the families of counterexamples that MPNNs are unable to distinguish from their distance matrices, by constructing families of novel and symmetric geometric graphs, to better understand the inherent limitations of MPNNs. We then propose $k$-DisGNNs, which can effectively exploit the rich geometry contained in the distance matrix. We demonstrate the high expressive power of $k$-DisGNNs from three perspectives: 1. They can learn high-order geometric information that cannot be captured by MPNNs. 2. They can unify some existing well-designed geometric models. 3. They are universal function approximators from geometric graphs to scalars (when $k\geq 2$) and vectors (when $k\geq 3$). Most importantly, we establish a connection between geometric deep learning (GDL) and traditional graph representation learning (GRL), showing that those highly expressive GNN models originally designed for GRL can also be applied to GDL with impressive performance, and that existing complex, equivariant models are not the only solution. Experiments verify our theory.
♻ ☆ Priors for symbolic regression GECCO
When choosing between competing symbolic models for a data set, a human will naturally prefer the "simpler" expression or the one which more closely resembles equations previously seen in a similar context. This suggests a non-uniform prior on functions, which is, however, rarely considered within a symbolic regression (SR) framework. In this paper we develop methods to incorporate detailed prior information on both functions and their parameters into SR. Our prior on the structure of a function is based on a $n$-gram language model, which is sensitive to the arrangement of operators relative to one another in addition to the frequency of occurrence of each operator. We also develop a formalism based on the Fractional Bayes Factor to treat numerical parameter priors in such a way that models may be fairly compared though the Bayesian evidence, and explicitly compare Bayesian, Minimum Description Length and heuristic methods for model selection. We demonstrate the performance of our priors relative to literature standards on benchmarks and a real-world dataset from the field of cosmology.
comment: 8+2 pages, 2 figures. Accepted for The Genetic and Evolutionary Computation Conference (GECCO) 2023 Workshop on Symbolic Regression
Multimedia 1
☆ JEPOO: Highly Accurate Joint Estimation of Pitch, Onset and Offset for Music Information Retrieval IJCAI 2023
Melody extraction is a core task in music information retrieval, and the estimation of pitch, onset and offset are key sub-tasks in melody extraction. Existing methods have limited accuracy, and work for only one type of data, either single-pitch or multipitch. In this paper, we propose a highly accurate method for joint estimation of pitch, onset and offset, named JEPOO. We address the challenges of joint learning optimization and handling both single-pitch and multi-pitch data through novel model design and a new optimization technique named Pareto modulated loss with loss weight regularization. This is the first method that can accurately handle both single-pitch and multi-pitch music data, and even a mix of them. A comprehensive experimental study on a wide range of real datasets shows that JEPOO outperforms state-ofthe-art methods by up to 10.6%, 8.3% and 10.3% for the prediction of Pitch, Onset and Offset, respectively, and JEPOO is robust for various types of data and instruments. The ablation study shows the effectiveness of each component of JEPOO.
comment: This paper has been accepted by IJCAI 2023; 11 pages, 6 figures
Computation and Language 143
☆ TopEx: Topic-based Explanations for Model Comparison ICLR
Meaningfully comparing language models is challenging with current explanation methods. Current explanations are overwhelming for humans due to large vocabularies or incomparable across models. We present TopEx, an explanation method that enables a level playing field for comparing language models via model-agnostic topics. We demonstrate how TopEx can identify similarities and differences between DistilRoBERTa and GPT-2 on a variety of NLP tasks.
comment: 7 pages, 4 figures, Accepted to ICLR Tiny Papers Track
☆ AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration
Large language models (LLMs) have shown excellent performance on various tasks, but the astronomical model size raises the hardware barrier for serving (memory size) and slows down token generation (memory bandwidth). In this paper, we propose Activation-aware Weight Quantization (AWQ), a hardware-friendly approach for LLM low-bit weight-only quantization. Our method is based on the observation that weights are not equally important: protecting only 1% of salient weights can greatly reduce quantization error. We then propose to search for the optimal per-channel scaling that protects the salient weights by observing the activation, not weights. AWQ does not rely on any backpropagation or reconstruction, so it can well preserve LLMs' generalization ability on different domains and modalities, without overfitting to the calibration set; it also does not rely on any data layout reordering, maintaining the hardware efficiency. AWQ outperforms existing work on various language modeling, common sense QA, and domain-specific benchmarks. Thanks to better generalization, it achieves excellent quantization performance for instruction-tuned LMs and, for the first time, multi-modal LMs. We also implement efficient tensor core kernels with reorder-free online dequantization to accelerate AWQ, achieving a 1.45x speedup over GPTQ and is 1.85x faster than the cuBLAS FP16 implementation. Our method provides a turn-key solution to compress LLMs to 3/4 bits for efficient deployment.
comment: Code available at: https://github.com/mit-han-lab/llm-awq
☆ EEL: Efficiently Encoding Lattices for Reranking ACL 2023
Standard decoding approaches for conditional text generation tasks typically search for an output hypothesis with high model probability, but this may not yield the best hypothesis according to human judgments of quality. Reranking to optimize for "downstream" metrics can better optimize for quality, but many metrics of interest are computed with pre-trained language models, which are slow to apply to large numbers of hypotheses. We explore an approach for reranking hypotheses by using Transformers to efficiently encode lattices of generated outputs, a method we call EEL. With a single Transformer pass over the entire lattice, we can approximately compute a contextualized representation of each token as if it were only part of a single hypothesis in isolation. We combine this approach with a new class of token-factored rerankers (TFRs) that allow for efficient extraction of high reranker-scoring hypotheses from the lattice. Empirically, our approach incurs minimal degradation error compared to the exponentially slower approach of encoding each hypothesis individually. When applying EEL with TFRs across three text generation tasks, our results show both substantial speedup compared to naive reranking and often better performance on downstream metrics than comparable approaches.
comment: ACL 2023 (16 pages), code available at https://github.com/PrasannS/eel-reranking
☆ Exposing Attention Glitches with Flip-Flop Language Modeling
Why do large language models sometimes output factual inaccuracies and exhibit erroneous reasoning? The brittleness of these models, particularly when executing long chains of reasoning, currently seems to be an inevitable price to pay for their advanced capabilities of coherently synthesizing knowledge, pragmatics, and abstract thought. Towards making sense of this fundamentally unsolved problem, this work identifies and analyzes the phenomenon of attention glitches, in which the Transformer architecture's inductive biases intermittently fail to capture robust reasoning. To isolate the issue, we introduce flip-flop language modeling (FFLM), a parametric family of synthetic benchmarks designed to probe the extrapolative behavior of neural language models. This simple generative task requires a model to copy binary symbols over long-range dependencies, ignoring the tokens in between. We find that Transformer FFLMs suffer from a long tail of sporadic reasoning errors, some of which we can eliminate using various regularization techniques. Our preliminary mechanistic analyses show why the remaining errors may be very difficult to diagnose and resolve. We hypothesize that attention glitches account for (some of) the closed-domain hallucinations in natural LLMs.
☆ AMR4NLI: Interpretable and robust NLI measures from semantic graphs
The task of natural language inference (NLI) asks whether a given premise (expressed in NL) entails a given NL hypothesis. NLI benchmarks contain human ratings of entailment, but the meaning relationships driving these ratings are not formalized. Can the underlying sentence pair relationships be made more explicit in an interpretable yet robust fashion? We compare semantic structures to represent premise and hypothesis, including sets of contextualized embeddings and semantic graphs (Abstract Meaning Representations), and measure whether the hypothesis is a semantic substructure of the premise, utilizing interpretable metrics. Our evaluation on three English benchmarks finds value in both contextualized embeddings and semantic graphs; moreover, they provide complementary signals, and can be leveraged together in a hybrid model.
comment: International Conference on Computational Semantics (IWCS 2023)
☆ "Let's not Quote out of Context": Unified Vision-Language Pretraining for Context Assisted Image Captioning
Well-formed context aware image captions and tags in enterprise content such as marketing material are critical to ensure their brand presence and content recall. Manual creation and updates to ensure the same is non trivial given the scale and the tedium towards this task. We propose a new unified Vision-Language (VL) model based on the One For All (OFA) model, with a focus on context-assisted image captioning where the caption is generated based on both the image and its context. Our approach aims to overcome the context-independent (image and text are treated independently) nature of the existing approaches. We exploit context by pretraining our model with datasets of three tasks: news image captioning where the news article is the context, contextual visual entailment, and keyword extraction from the context. The second pretraining task is a new VL task, and we construct and release two datasets for the task with 1.1M and 2.2K data instances. Our system achieves state-of-the-art results with an improvement of up to 8.34 CIDEr score on the benchmark news image captioning datasets. To the best of our knowledge, ours is the first effort at incorporating contextual information in pretraining the models for the VL tasks.
☆ ACLM: A Selective-Denoising based Generative Data Augmentation Approach for Low-Resource Complex NER ACL 2023
Complex Named Entity Recognition (NER) is the task of detecting linguistically complex named entities in low-context text. In this paper, we present ACLM Attention-map aware keyword selection for Conditional Language Model fine-tuning), a novel data augmentation approach based on conditional generation to address the data scarcity problem in low-resource complex NER. ACLM alleviates the context-entity mismatch issue, a problem existing NER data augmentation techniques suffer from and often generates incoherent augmentations by placing complex named entities in the wrong context. ACLM builds on BART and is optimized on a novel text reconstruction or denoising task - we use selective masking (aided by attention maps) to retain the named entities and certain keywords in the input sentence that provide contextually relevant additional knowledge or hints about the named entities. Compared with other data augmentation strategies, ACLM can generate more diverse and coherent augmentations preserving the true word sense of complex entities in the sentence. We demonstrate the effectiveness of ACLM both qualitatively and quantitatively on monolingual, cross-lingual, and multilingual complex NER across various low-resource settings. ACLM outperforms all our neural baselines by a significant margin (1%-36%). In addition, we demonstrate the application of ACLM to other domains that suffer from data scarcity (e.g., biomedical). In practice, ACLM generates more effective and factual augmentations for these domains than prior methods. Code: https://github.com/Sreyan88/ACLM
comment: ACL 2023 Main Conference
☆ Minding Language Models' (Lack of) Theory of Mind: A Plug-and-Play Multi-Character Belief Tracker
Theory of Mind (ToM)$\unicode{x2014}$the ability to reason about the mental states of other people$\unicode{x2014}$is a key element of our social intelligence. Yet, despite their ever more impressive performance, large-scale neural language models still lack basic theory of mind capabilities out-of-the-box. We posit that simply scaling up models will not imbue them with theory of mind due to the inherently symbolic and implicit nature of the phenomenon, and instead investigate an alternative: can we design a decoding-time algorithm that enhances theory of mind of off-the-shelf neural language models without explicit supervision? We present SymbolicToM, a plug-and-play approach to reason about the belief states of multiple characters in reading comprehension tasks via explicit symbolic representation. More concretely, our approach tracks each entity's beliefs, their estimation of other entities' beliefs, and higher-order levels of reasoning, all through graphical representations, allowing for more precise and interpretable reasoning than previous approaches. Empirical results on the well-known ToMi benchmark (Le et al., 2019) demonstrate that SymbolicToM dramatically enhances off-the-shelf neural networks' theory of mind in a zero-shot setting while showing robust out-of-distribution performance compared to supervised baselines. Our work also reveals spurious patterns in existing theory of mind benchmarks, emphasizing the importance of out-of-distribution evaluation and methods that do not overfit a particular dataset.
☆ T2IAT: Measuring Valence and Stereotypical Biases in Text-to-Image Generation ACL 2023
Warning: This paper contains several contents that may be toxic, harmful, or offensive. In the last few years, text-to-image generative models have gained remarkable success in generating images with unprecedented quality accompanied by a breakthrough of inference speed. Despite their rapid progress, human biases that manifest in the training examples, particularly with regard to common stereotypical biases, like gender and skin tone, still have been found in these generative models. In this work, we seek to measure more complex human biases exist in the task of text-to-image generations. Inspired by the well-known Implicit Association Test (IAT) from social psychology, we propose a novel Text-to-Image Association Test (T2IAT) framework that quantifies the implicit stereotypes between concepts and valence, and those in the images. We replicate the previously documented bias tests on generative models, including morally neutral tests on flowers and insects as well as demographic stereotypical tests on diverse social attributes. The results of these experiments demonstrate the presence of complex stereotypical behaviors in image generations.
comment: ACL 2023
☆ LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Day
Conversational generative AI has demonstrated remarkable promise for empowering biomedical practitioners, but current investigations focus on unimodal text. Multimodal conversational AI has seen rapid progress by leveraging billions of image-text pairs from the public web, but such general-domain vision-language models still lack sophistication in understanding and conversing about biomedical images. In this paper, we propose a cost-efficient approach for training a vision-language conversational assistant that can answer open-ended research questions of biomedical images. The key idea is to leverage a large-scale, broad-coverage biomedical figure-caption dataset extracted from PubMed Central, use GPT-4 to self-instruct open-ended instruction-following data from the captions, and then fine-tune a large general-domain vision-language model using a novel curriculum learning method. Specifically, the model first learns to align biomedical vocabulary using the figure-caption pairs as is, then learns to master open-ended conversational semantics using GPT-4 generated instruction-following data, broadly mimicking how a layperson gradually acquires biomedical knowledge. This enables us to train a Large Language and Vision Assistant for BioMedicine (LLaVA-Med) in less than 15 hours (with eight A100s). LLaVA-Med exhibits excellent multimodal conversational capability and can follow open-ended instruction to assist with inquiries about a biomedical image. On three standard biomedical visual question answering datasets, LLaVA-Med outperforms previous supervised state-of-the-art on certain metrics. To facilitate biomedical multimodal research, we will release our instruction-following data and the LLaVA-Med model.
comment: 17 pages; Website: https://aka.ms/llava-med
☆ OpenPI-C: A Better Benchmark and Stronger Baseline for Open-Vocabulary State Tracking ACL 2023
Open-vocabulary state tracking is a more practical version of state tracking that aims to track state changes of entities throughout a process without restricting the state space and entity space. OpenPI is to date the only dataset annotated for open-vocabulary state tracking. However, we identify issues with the dataset quality and evaluation metric. For the dataset, we categorize 3 types of problems on the procedure level, step level and state change level respectively, and build a clean dataset OpenPI-C using multiple rounds of human judgment. For the evaluation metric, we propose a cluster-based metric to fix the original metric's preference for repetition. Model-wise, we enhance the seq2seq generation baseline by reinstating two key properties for state tracking: temporal dependency and entity awareness. The state of the world after an action is inherently dependent on the previous state. We model this dependency through a dynamic memory bank and allow the model to attend to the memory slots during decoding. On the other hand, the state of the world is naturally a union of the states of involved entities. Since the entities are unknown in the open-vocabulary setting, we propose a two-stage model that refines the state change prediction conditioned on entities predicted from the first stage. Empirical results show the effectiveness of our proposed model especially on the cluster-based metric. The code and data are released at https://github.com/shirley-wu/openpi-c
comment: ACL 2023 findings
☆ A Transformer-based representation-learning model with unified processing of multimodal input for clinical diagnostics
During the diagnostic process, clinicians leverage multimodal information, such as chief complaints, medical images, and laboratory-test results. Deep-learning models for aiding diagnosis have yet to meet this requirement. Here we report a Transformer-based representation-learning model as a clinical diagnostic aid that processes multimodal input in a unified manner. Rather than learning modality-specific features, the model uses embedding layers to convert images and unstructured and structured text into visual tokens and text tokens, and bidirectional blocks with intramodal and intermodal attention to learn a holistic representation of radiographs, the unstructured chief complaint and clinical history, structured clinical information such as laboratory-test results and patient demographic information. The unified model outperformed an image-only model and non-unified multimodal diagnosis models in the identification of pulmonary diseases (by 12% and 9%, respectively) and in the prediction of adverse clinical outcomes in patients with COVID-19 (by 29% and 7%, respectively). Leveraging unified multimodal Transformer-based models may help streamline triage of patients and facilitate the clinical decision process.
comment: Accepted by Nature Biomedical Engineering
☆ Adversarial learning of neural user simulators for dialogue policy optimisation
Reinforcement learning based dialogue policies are typically trained in interaction with a user simulator. To obtain an effective and robust policy, this simulator should generate user behaviour that is both realistic and varied. Current data-driven simulators are trained to accurately model the user behaviour in a dialogue corpus. We propose an alternative method using adversarial learning, with the aim to simulate realistic user behaviour with more variation. We train and evaluate several simulators on a corpus of restaurant search dialogues, and then use them to train dialogue system policies. In policy cross-evaluation experiments we demonstrate that an adversarially trained simulator produces policies with 8.3% higher success rate than those trained with a maximum likelihood simulator. Subjective results from a crowd-sourced dialogue system user evaluation confirm the effectiveness of adversarially training user simulators.
comment: UK Speech 2023
☆ Zero and Few-shot Semantic Parsing with Ambiguous Inputs
Despite the ubiquity of ambiguity in natural language, it is often ignored or deliberately removed in semantic parsing tasks, which generally assume that a given surface form has only one correct logical form. We attempt to address this shortcoming by introducing AmP, a framework, dataset, and challenge for parsing with linguistic ambiguity. We define templates and generate data for five well-documented linguistic ambiguities. Using AmP, we investigate how several few-shot semantic parsing systems handle ambiguity, introducing three new metrics. We find that large pre-trained models perform poorly at capturing the distribution of possible meanings without deliberate instruction. However, models are able to capture distribution well when ambiguity is attested in their inputs. These results motivate a call for ambiguity to be explicitly included in semantic parsing, and promotes considering the distribution of possible outputs when evaluating semantic parsing systems.
comment: 15 pages
☆ Adaptive Contextual Biasing for Transducer Based Streaming Speech Recognition
By incorporating additional contextual information, deep biasing methods have emerged as a promising solution for speech recognition of personalized words. However, for real-world voice assistants, always biasing on such personalized words with high prediction scores can significantly degrade the performance of recognizing common words. To address this issue, we propose an adaptive contextual biasing method based on Context-Aware Transformer Transducer (CATT) that utilizes the biased encoder and predictor embeddings to perform streaming prediction of contextual phrase occurrences. Such prediction is then used to dynamically switch the bias list on and off, enabling the model to adapt to both personalized and common scenarios. Experiments on Librispeech and internal voice assistant datasets show that our approach can achieve up to 6.7% and 20.7% relative reduction in WER and CER compared to the baseline respectively, mitigating up to 96.7% and 84.9% of the relative WER and CER increase for common cases. Furthermore, our approach has a minimal performance impact in personalized scenarios while maintaining a streaming inference pipeline with negligible RTF increase.
☆ Birth of a Transformer: A Memory Viewpoint
Large language models based on transformers have achieved great empirical successes. However, as they are deployed more widely, there is a growing need to better understand their internal mechanisms in order to make them more reliable. These models appear to store vast amounts of knowledge from their training data, and to adapt quickly to new information provided in their context or prompt. We study how transformers balance these two types of knowledge by considering a synthetic setup where tokens are generated from either global or context-specific bigram distributions. By a careful empirical analysis of the training process on a simplified two-layer transformer, we illustrate the fast learning of global bigrams and the slower development of an "induction head" mechanism for the in-context bigrams. We highlight the role of weight matrices as associative memories, provide theoretical insights on how gradients enable their learning during training, and study the role of data-distributional properties.
☆ Modeling and Analyzing Scorer Preferences in Short-Answer Math Questions
Automated scoring of student responses to open-ended questions, including short-answer questions, has great potential to scale to a large number of responses. Recent approaches for automated scoring rely on supervised learning, i.e., training classifiers or fine-tuning language models on a small number of responses with human-provided score labels. However, since scoring is a subjective process, these human scores are noisy and can be highly variable, depending on the scorer. In this paper, we investigate a collection of models that account for the individual preferences and tendencies of each human scorer in the automated scoring task. We apply these models to a short-answer math response dataset where each response is scored (often differently) by multiple different human scorers. We conduct quantitative experiments to show that our scorer models lead to improved automated scoring accuracy. We also conduct quantitative experiments and case studies to analyze the individual preferences and tendencies of scorers. We found that scorers can be grouped into several obvious clusters, with each cluster having distinct features, and analyzed them in detail.
comment: Accepted to 16th International Conference on Educational Data Mining (EDM 2023)
☆ Improved Cross-Lingual Transfer Learning For Automatic Speech Translation
Research in multilingual speech-to-text translation is topical. Having a single model that supports multiple translation tasks is desirable. The goal of this work it to improve cross-lingual transfer learning in multilingual speech-to-text translation via semantic knowledge distillation. We show that by initializing the encoder of the encoder-decoder sequence-to-sequence translation model with SAMU-XLS-R, a multilingual speech transformer encoder trained using multi-modal (speech-text) semantic knowledge distillation, we achieve significantly better cross-lingual task knowledge transfer than the baseline XLS-R, a multilingual speech transformer encoder trained via self-supervised learning. We demonstrate the effectiveness of our approach on two popular datasets, namely, CoVoST-2 and Europarl. On the 21 translation tasks of the CoVoST-2 benchmark, we achieve an average improvement of 12.8 BLEU points over the baselines. In the zero-shot translation scenario, we achieve an average gain of 18.8 and 11.9 average BLEU points on unseen medium and low-resource languages. We make similar observations on Europarl speech translation benchmark.
☆ Interpretable Math Word Problem Solution Generation Via Step-by-step Planning ACL 2023
Solutions to math word problems (MWPs) with step-by-step explanations are valuable, especially in education, to help students better comprehend problem-solving strategies. Most existing approaches only focus on obtaining the final correct answer. A few recent approaches leverage intermediate solution steps to improve final answer correctness but often cannot generate coherent steps with a clear solution strategy. Contrary to existing work, we focus on improving the correctness and coherence of the intermediate solutions steps. We propose a step-by-step planning approach for intermediate solution generation, which strategically plans the generation of the next solution step based on the MWP and the previous solution steps. Our approach first plans the next step by predicting the necessary math operation needed to proceed, given history steps, then generates the next step, token-by-token, by prompting a language model with the predicted math operation. Experiments on the GSM8K dataset demonstrate that our approach improves the accuracy and interpretability of the solution on both automatic metrics and human evaluation.
comment: Accepted to The 61st Annual Meeting of the Association for Computational Linguistics (ACL 2023)
☆ In-Context Learning User Simulators for Task-Oriented Dialog Systems
This paper presents a novel application of large language models in user simulation for task-oriented dialog systems, specifically focusing on an in-context learning approach. By harnessing the power of these models, the proposed approach generates diverse utterances based on user goals and limited dialog examples. Unlike traditional simulators, this method eliminates the need for labor-intensive rule definition or extensive annotated data, making it more efficient and accessible. Additionally, an error analysis of the interaction between the user simulator and dialog system uncovers common mistakes, providing valuable insights into areas that require improvement. Our implementation is available at https://github.com/telepathylabsai/prompt-based-user-simulator.
☆ Topic-Guided Sampling For Data-Efficient Multi-Domain Stance Detection ACL 2023
Stance Detection is concerned with identifying the attitudes expressed by an author towards a target of interest. This task spans a variety of domains ranging from social media opinion identification to detecting the stance for a legal claim. However, the framing of the task varies within these domains, in terms of the data collection protocol, the label dictionary and the number of available annotations. Furthermore, these stance annotations are significantly imbalanced on a per-topic and inter-topic basis. These make multi-domain stance detection a challenging task, requiring standardization and domain adaptation. To overcome this challenge, we propose $\textbf{T}$opic $\textbf{E}$fficient $\textbf{St}$anc$\textbf{E}$ $\textbf{D}$etection (TESTED), consisting of a topic-guided diversity sampling technique and a contrastive objective that is used for fine-tuning a stance classifier. We evaluate the method on an existing benchmark of $16$ datasets with in-domain, i.e. all topics seen and out-of-domain, i.e. unseen topics, experiments. The results show that our method outperforms the state-of-the-art with an average of $3.5$ F1 points increase in-domain, and is more generalizable with an averaged increase of $10.2$ F1 on out-of-domain evaluation while using $\leq10\%$ of the training data. We show that our sampling technique mitigates both inter- and per-topic class imbalances. Finally, our analysis demonstrates that the contrastive learning objective allows the model a more pronounced segmentation of samples with varying labels.
comment: ACL 2023 (Oral)
☆ Enhancing the Unified Streaming and Non-streaming Model with Contrastive Learning INTERSPEECH 2023
The unified streaming and non-streaming speech recognition model has achieved great success due to its comprehensive capabilities. In this paper, we propose to improve the accuracy of the unified model by bridging the inherent representation gap between the streaming and non-streaming modes with a contrastive objective. Specifically, the top-layer hidden representation at the same frame of the streaming and non-streaming modes are regarded as a positive pair, encouraging the representation of the streaming mode close to its non-streaming counterpart. The multiple negative samples are randomly selected from the rest frames of the same sample under the non-streaming mode. Experimental results demonstrate that the proposed method achieves consistent improvements toward the unified model in both streaming and non-streaming modes. Our method achieves CER of 4.66% in the streaming mode and CER of 4.31% in the non-streaming mode, which sets a new state-of-the-art on the AISHELL-1 benchmark.
comment: Accepted by INTERSPEECH 2023
☆ Differentiable Tree Operations Promote Compositional Generalization ICML 2023
In the context of structure-to-structure transformation tasks, learning sequences of discrete symbolic operations poses significant challenges due to their non-differentiability. To facilitate the learning of these symbolic sequences, we introduce a differentiable tree interpreter that compiles high-level symbolic tree operations into subsymbolic matrix operations on tensors. We present a novel Differentiable Tree Machine (DTM) architecture that integrates our interpreter with an external memory and an agent that learns to sequentially select tree operations to execute the target transformation in an end-to-end manner. With respect to out-of-distribution compositional generalization on synthetic semantic parsing and language generation tasks, DTM achieves 100% while existing baselines such as Transformer, Tree Transformer, LSTM, and Tree2Tree LSTM achieve less than 30%. DTM remains highly interpretable in addition to its perfect performance.
comment: ICML 2023. Code available at https://github.com/psoulos/dtm
☆ Column Type Annotation using ChatGPT
Column type annotation is the task of annotating the columns of a relational table with the semantic type of the values contained in each column. Column type annotation is a crucial pre-processing step for data search and integration in the context of data lakes. State-of-the-art column type annotation methods either rely on matching table columns to properties of a knowledge graph or fine-tune pre-trained language models such as BERT for the column type annotation task. In this work, we take a different approach and explore using ChatGPT for column type annotation. We evaluate different prompt designs in zero- and few-shot settings and experiment with providing task definitions and detailed instructions to the model. We further implement a two-step table annotation pipeline which first determines the class of the entities described in the table and depending on this class asks ChatGPT to annotate columns using only the relevant subset of the overall vocabulary. Using instructions as well as the two-step pipeline, ChatGPT reaches F1 scores of over 85% in zero- and one-shot setups. To reach a similar F1 score a RoBERTa model needs to be fine-tuned with 300 examples. This comparison shows that ChatGPT is able deliver competitive results for the column type annotation task given no or only a minimal amount of task-specific demonstrations.
☆ ReFACT: Updating Text-to-Image Models by Editing the Text Encoder
Text-to-image models are trained on extensive amounts of data, leading them to implicitly encode factual knowledge within their parameters. While some facts are useful, others may be incorrect or become outdated (e.g., the current President of the United States). We introduce ReFACT, a novel approach for editing factual knowledge in text-to-image generative models. ReFACT updates the weights of a specific layer in the text encoder, only modifying a tiny portion of the model's parameters, and leaving the rest of the model unaffected. We empirically evaluate ReFACT on an existing benchmark, alongside RoAD, a newly curated dataset. ReFACT achieves superior performance in terms of generalization to related concepts while preserving unrelated concepts. Furthermore, ReFACT maintains image generation quality, making it a valuable tool for updating and correcting factual information in text-to-image models.
☆ Boosting the Performance of Transformer Architectures for Semantic Textual Similarity
Semantic textual similarity is the task of estimating the similarity between the meaning of two texts. In this paper, we fine-tune transformer architectures for semantic textual similarity on the Semantic Textual Similarity Benchmark by tuning the model partially and then end-to-end. We experiment with BERT, RoBERTa, and DeBERTaV3 cross-encoders by approaching the problem as a binary classification task or a regression task. We combine the outputs of the transformer models and use handmade features as inputs for boosting algorithms. Due to worse test set results coupled with improvements on the validation set, we experiment with different dataset splits to further investigate this occurrence. We also provide an error analysis, focused on the edges of the prediction range.
comment: 6 pages, 2 figures, 12 tables
☆ How Generative Spoken Language Modeling Encodes Noisy Speech: Investigation from Phonetics to Syntactics INTERSPEECH 2023
We examine the speech modeling potential of generative spoken language modeling (GSLM), which involves using learned symbols derived from data rather than phonemes for speech analysis and synthesis. Since GSLM facilitates textless spoken language processing, exploring its effectiveness is critical for paving the way for novel paradigms in spoken-language processing. This paper presents the findings of GSLM's encoding and decoding effectiveness at the spoken-language and speech levels. Through speech resynthesis experiments, we revealed that resynthesis errors occur at the levels ranging from phonology to syntactics and GSLM frequently resynthesizes natural but content-altered speech.
comment: Accepted to INTERSPEECH 2023
☆ Towards Argument-Aware Abstractive Summarization of Long Legal Opinions with Summary Reranking
We propose a simple approach for the abstractive summarization of long legal opinions that considers the argument structure of the document. Legal opinions often contain complex and nuanced argumentation, making it challenging to generate a concise summary that accurately captures the main points of the legal opinion. Our approach involves using argument role information to generate multiple candidate summaries, then reranking these candidates based on alignment with the document's argument structure. We demonstrate the effectiveness of our approach on a dataset of long legal opinions and show that it outperforms several strong baselines.
☆ Predicting the Quality of Revisions in Argumentative Writing ACL
The ability to revise in response to feedback is critical to students' writing success. In the case of argument writing in specific, identifying whether an argument revision (AR) is successful or not is a complex problem because AR quality is dependent on the overall content of an argument. For example, adding the same evidence sentence could strengthen or weaken existing claims in different argument contexts (ACs). To address this issue we developed Chain-of-Thought prompts to facilitate ChatGPT-generated ACs for AR quality predictions. The experiments on two corpora, our annotated elementary essays and existing college essays benchmark, demonstrate the superiority of the proposed ACs over baselines.
comment: In The 18th BEA Workshop, held in conjunction with The Association for Computational Linguistics (ACL), July 2023
☆ Automatic Glossary of Clinical Terminology: a Large-Scale Dictionary of Biomedical Definitions Generated from Ontological Knowledge
Background: More than 400,000 biomedical concepts and some of their relationships are contained in SnomedCT, a comprehensive biomedical ontology. However, their concept names are not always readily interpretable by non-experts, or patients looking at their own electronic health records (EHR). Clear definitions or descriptions in understandable language are often not available. Therefore, generating human-readable definitions for biomedical concepts might help make the information they encode more accessible and understandable to a wider public. Objective: In this article, we introduce the Automatic Glossary of Clinical Terminology (AGCT), a large-scale biomedical dictionary of clinical concepts generated using high-quality information extracted from the biomedical knowledge contained in SnomedCT. Methods: We generate a novel definition for every SnomedCT concept, after prompting the OpenAI Turbo model, a variant of GPT 3.5, using a high-quality verbalization of the SnomedCT relationships of the to-be-defined concept. A significant subset of the generated definitions was subsequently judged by NLP researchers with biomedical expertise on 5-point scales along the following three axes: factuality, insight, and fluency. Results: AGCT contains 422,070 computer-generated definitions for SnomedCT concepts, covering various domains such as diseases, procedures, drugs, and anatomy. The average length of the definitions is 49 words. The definitions were assigned average scores of over 4.5 out of 5 on all three axes, indicating a majority of factual, insightful, and fluent definitions. Conclusion: AGCT is a novel and valuable resource for biomedical tasks that require human-readable definitions for SnomedCT concepts. It can also serve as a base for developing robust biomedical retrieval models or other applications that leverage natural language understanding of biomedical knowledge.
comment: Accepted at the BioNLP 2023 workshop
☆ Improving Polish to English Neural Machine Translation with Transfer Learning: Effects of Data Volume and Language Similarity
This paper investigates the impact of data volume and the use of similar languages on transfer learning in a machine translation task. We find out that having more data generally leads to better performance, as it allows the model to learn more patterns and generalizations from the data. However, related languages can also be particularly effective when there is limited data available for a specific language pair, as the model can leverage the similarities between the languages to improve performance. To demonstrate, we fine-tune mBART model for a Polish-English translation task using the OPUS-100 dataset. We evaluate the performance of the model under various transfer learning configurations, including different transfer source languages and different shot levels for Polish, and report the results. Our experiments show that a combination of related languages and larger amounts of data outperforms the model trained on related languages or larger amounts of data alone. Additionally, we show the importance of related languages in zero-shot and few-shot configurations.
☆ Explanation Graph Generation via Generative Pre-training over Synthetic Graphs ACL23
The generation of explanation graphs is a significant task that aims to produce explanation graphs in response to user input, revealing the internal reasoning process. This task is challenging due to the significant discrepancy between unstructured user queries and structured explanation graphs. Current research commonly fine-tunes a text-based pre-trained language model on a small downstream dataset that is annotated with labeled graphs. However, due to the limited scale of available datasets, this approach may prove to be insufficient in bridging the gap between natural language text and structured graphs. In this paper, to alleviate the above limitations, we propose a novel pre-trained framework EG3P(for Explanation Graph Generation via Generative Pre-training over synthetic graphs) for the explanation graph generation task. Specifically, we first propose a text-to-graph generative task to pre-train the model with the goal of bridging the text-graph gap. Additionally, we propose an automatic corpus synthesis strategy for synthesizing a large scale of high-quality corpus, reducing the reliance on costly manual annotation methods. Experimental results on ExplaGraphs show the effectiveness of EG3P that our model surpasses all baseline systems with remarkable margins. Besides, further analysis demonstrates that EG3P is able to generate better explanation graphs on actual reasoning tasks such as CommonsenseQA and OpenbookQA.
comment: Accepted by ACL23-Findings
☆ Contextual Distortion Reveals Constituency: Masked Language Models are Implicit Parsers ACL 2023
Recent advancements in pre-trained language models (PLMs) have demonstrated that these models possess some degree of syntactic awareness. To leverage this knowledge, we propose a novel chart-based method for extracting parse trees from masked language models (LMs) without the need to train separate parsers. Our method computes a score for each span based on the distortion of contextual representations resulting from linguistic perturbations. We design a set of perturbations motivated by the linguistic concept of constituency tests, and use these to score each span by aggregating the distortion scores. To produce a parse tree, we use chart parsing to find the tree with the minimum score. Our method consistently outperforms previous state-of-the-art methods on English with masked LMs, and also demonstrates superior performance in a multilingual setting, outperforming the state of the art in 6 out of 8 languages. Notably, although our method does not involve parameter updates or extensive hyperparameter search, its performance can even surpass some unsupervised parsing methods that require fine-tuning. Our analysis highlights that the distortion of contextual representation resulting from syntactic perturbation can serve as an effective indicator of constituency across languages.
comment: Accepted by ACL 2023
☆ Being Right for Whose Right Reasons? ACL 2023
Explainability methods are used to benchmark the extent to which model predictions align with human rationales i.e., are 'right for the right reasons'. Previous work has failed to acknowledge, however, that what counts as a rationale is sometimes subjective. This paper presents what we think is a first of its kind, a collection of human rationale annotations augmented with the annotators demographic information. We cover three datasets spanning sentiment analysis and common-sense reasoning, and six demographic groups (balanced across age and ethnicity). Such data enables us to ask both what demographics our predictions align with and whose reasoning patterns our models' rationales align with. We find systematic inter-group annotator disagreement and show how 16 Transformer-based models align better with rationales provided by certain demographic groups: We find that models are biased towards aligning best with older and/or white annotators. We zoom in on the effects of model size and model distillation, finding -- contrary to our expectations -- negative correlations between model size and rationale agreement as well as no evidence that either model size or model distillation improves fairness.
comment: To be published in ACL 2023
ReviewerGPT? An Exploratory Study on Using Large Language Models for Paper Reviewing
Given the rapid ascent of large language models (LLMs), we study the question: (How) can large language models help in reviewing of scientific papers or proposals? We first conduct some pilot studies where we find that (i) GPT-4 outperforms other LLMs (Bard, Vicuna, Koala, Alpaca, LLaMa, Dolly, OpenAssistant, StableLM), and (ii) prompting with a specific question (e.g., to identify errors) outperforms prompting to simply write a review. With these insights, we study the use of LLMs (specifically, GPT-4) for three tasks: 1. Identifying errors: We construct 13 short computer science papers each with a deliberately inserted error, and ask the LLM to check for the correctness of these papers. We observe that the LLM finds errors in 7 of them, spanning both mathematical and conceptual errors. 2. Verifying checklists: We task the LLM to verify 16 closed-ended checklist questions in the respective sections of 15 NeurIPS 2022 papers. We find that across 119 {checklist question, paper} pairs, the LLM had an 86.6% accuracy. 3. Choosing the "better" paper: We generate 10 pairs of abstracts, deliberately designing each pair in such a way that one abstract was clearly superior than the other. The LLM, however, struggled to discern these relatively straightforward distinctions accurately, committing errors in its evaluations for 6 out of the 10 pairs. Based on these experiments, we think that LLMs have a promising use as reviewing assistants for specific reviewing tasks, but not (yet) for complete evaluations of papers or proposals.
☆ Effective Structured Prompting by Meta-Learning and Representative Verbalizer ICML 2023
Prompt tuning for pre-trained masked language models (MLM) has shown promising performance in natural language processing tasks with few labeled examples. It tunes a prompt for the downstream task, and a verbalizer is used to bridge the predicted token and label prediction. Due to the limited training data, prompt initialization is crucial for prompt tuning. Recently, MetaPrompting (Hou et al., 2022) uses meta-learning to learn a shared initialization for all task-specific prompts. However, a single initialization is insufficient to obtain good prompts for all tasks and samples when the tasks are complex. Moreover, MetaPrompting requires tuning the whole MLM, causing a heavy burden on computation and memory as the MLM is usually large. To address these issues, we use a prompt pool to extract more task knowledge and construct instance-dependent prompts via attention. We further propose a novel soft verbalizer (RepVerb) which constructs label embedding from feature embeddings directly. Combining meta-learning the prompt pool and RepVerb, we propose MetaPrompter for effective structured prompting. MetaPrompter is parameter-efficient as only the pool is required to be tuned. Experimental results demonstrate that MetaPrompter performs better than the recent state-of-the-arts and RepVerb outperforms existing soft verbalizers.
comment: Accepted at ICML 2023
☆ Enhancing Programming eTextbooks with ChatGPT Generated Counterfactual-Thinking-Inspired Questions
Digital textbooks have become an integral part of everyday learning tasks. In this work, we consider the use of digital textbooks for programming classes. Generally, students struggle with utilizing textbooks on programming to the maximum, with a possible reason being that the example programs provided as illustration of concepts in these textbooks don't offer sufficient interactivity for students, and thereby not sufficiently motivating to explore or understand these programming examples better. In our work, we explore the idea of enhancing the navigability of intelligent textbooks with the use of ``counterfactual'' questions, to make students think critically about these programs and enhance possible program comprehension. Inspired from previous works on nudging students on counter factual thinking, we present the possibility to enhance digital textbooks with questions generated using GPT.
comment: Paper Under Review
☆ Chain-Of-Thought Prompting Under Streaming Batch: A Case Study
Recently, Large Language Models (LLMs) have demonstrated remarkable capabilities. Chain-of-Thought (CoT) has been proposed as a way of assisting LLMs in performing complex reasoning. However, developing effective prompts can be a challenging and labor-intensive task. Many studies come out of some way to automatically construct CoT from test data. Most of them assume that all test data is visible before testing and only select a small subset to generate rationales, which is an unrealistic assumption. In this paper, we present a case study on how to construct and optimize chain-of-thought prompting using batch data in streaming settings.
☆ A Call for Standardization and Validation of Text Style Transfer Evaluation ACL 2023
Text Style Transfer (TST) evaluation is, in practice, inconsistent. Therefore, we conduct a meta-analysis on human and automated TST evaluation and experimentation that thoroughly examines existing literature in the field. The meta-analysis reveals a substantial standardization gap in human and automated evaluation. In addition, we also find a validation gap: only few automated metrics have been validated using human experiments. To this end, we thoroughly scrutinize both the standardization and validation gap and reveal the resulting pitfalls. This work also paves the way to close the standardization and validation gap in TST evaluation by calling out requirements to be met by future research.
comment: Accepted to Findings of ACL 2023
☆ The Effects of Input Type and Pronunciation Dictionary Usage in Transfer Learning for Low-Resource Text-to-Speech INTERSPEECH 2023
We compare phone labels and articulatory features as input for cross-lingual transfer learning in text-to-speech (TTS) for low-resource languages (LRLs). Experiments with FastSpeech 2 and the LRL West Frisian show that using articulatory features outperformed using phone labels in both intelligibility and naturalness. For LRLs without pronunciation dictionaries, we propose two novel approaches: a) using a massively multilingual model to convert grapheme-to-phone (G2P) in both training and synthesizing, and b) using a universal phone recognizer to create a makeshift dictionary. Results show that the G2P approach performs largely on par with using a ground-truth dictionary and the phone recognition approach, while performing generally worse, remains a viable option for LRLs less suitable for the G2P approach. Within each approach, using articulatory features as input outperforms using phone labels.
comment: Accepted at INTERSPEECH 2023
☆ Layout and Task Aware Instruction Prompt for Zero-shot Document Image Question Answering
The pre-training-fine-tuning paradigm based on layout-aware multimodal pre-trained models has achieved significant progress on document image question answering. However, domain pre-training and task fine-tuning for additional visual, layout, and task modules prevent them from directly utilizing off-the-shelf instruction-tuning language foundation models, which have recently shown promising potential in zero-shot learning. Contrary to aligning language models to the domain of document image question answering, we align document image question answering to off-the-shell instruction-tuning language foundation models to utilize their zero-shot capability. Specifically, we propose layout and task aware instruction prompt called LATIN-Prompt, which consists of layout-aware document content and task-aware descriptions. The former recovers the layout information among text segments from OCR tools by appropriate spaces and line breaks. The latter ensures that the model generates answers that meet the requirements, especially format requirements, through a detailed description of task. Experimental results on three benchmarks show that LATIN-Prompt can improve the zero-shot performance of instruction-tuning language foundation models on document image question answering and help them achieve comparable levels to SOTAs based on the pre-training-fine-tuning paradigm. Quantitative analysis and qualitative analysis demonstrate the effectiveness of LATIN-Prompt. We provide the code in supplementary and will release the code to facilitate future research.
comment: Code is available at https://github.com/WenjinW/LATIN-Prompt
☆ MEWL: Few-shot multimodal word learning with referential uncertainty ICML 2023
Without explicit feedback, humans can rapidly learn the meaning of words. Children can acquire a new word after just a few passive exposures, a process known as fast mapping. This word learning capability is believed to be the most fundamental building block of multimodal understanding and reasoning. Despite recent advancements in multimodal learning, a systematic and rigorous evaluation is still missing for human-like word learning in machines. To fill in this gap, we introduce the MachinE Word Learning (MEWL) benchmark to assess how machines learn word meaning in grounded visual scenes. MEWL covers human's core cognitive toolkits in word learning: cross-situational reasoning, bootstrapping, and pragmatic learning. Specifically, MEWL is a few-shot benchmark suite consisting of nine tasks for probing various word learning capabilities. These tasks are carefully designed to be aligned with the children's core abilities in word learning and echo the theories in the developmental literature. By evaluating multimodal and unimodal agents' performance with a comparative analysis of human performance, we notice a sharp divergence in human and machine word learning. We further discuss these differences between humans and machines and call for human-like few-shot word learning in machines.
comment: Accepted at ICML 2023
☆ Revisiting Event Argument Extraction: Can EAE Models Learn Better When Being Aware of Event Co-occurrences? ACL 2023
Event co-occurrences have been proved effective for event extraction (EE) in previous studies, but have not been considered for event argument extraction (EAE) recently. In this paper, we try to fill this gap between EE research and EAE research, by highlighting the question that ``Can EAE models learn better when being aware of event co-occurrences?''. To answer this question, we reformulate EAE as a problem of table generation and extend a SOTA prompt-based EAE model into a non-autoregressive generation framework, called TabEAE, which is able to extract the arguments of multiple events in parallel. Under this framework, we experiment with 3 different training-inference schemes on 4 datasets (ACE05, RAMS, WikiEvents and MLEE) and discover that via training the model to extract all events in parallel, it can better distinguish the semantic boundary of each event and its ability to extract single event gets substantially improved. Experimental results show that our method achieves new state-of-the-art performance on the 4 datasets. Our code is avilable at https://github.com/Stardust-hyx/TabEAE.
comment: Accepted to ACL 2023 main conference
☆ Inspecting Spoken Language Understanding from Kids for Basic Math Learning at Home ACL 2023
Enriching the quality of early childhood education with interactive math learning at home systems, empowered by recent advances in conversational AI technologies, is slowly becoming a reality. With this motivation, we implement a multimodal dialogue system to support play-based learning experiences at home, guiding kids to master basic math concepts. This work explores Spoken Language Understanding (SLU) pipeline within a task-oriented dialogue system developed for Kid Space, with cascading Automatic Speech Recognition (ASR) and Natural Language Understanding (NLU) components evaluated on our home deployment data with kids going through gamified math learning activities. We validate the advantages of a multi-task architecture for NLU and experiment with a diverse set of pretrained language representations for Intent Recognition and Entity Extraction tasks in the math learning domain. To recognize kids' speech in realistic home environments, we investigate several ASR systems, including the commercial Google Cloud and the latest open-source Whisper solutions with varying model sizes. We evaluate the SLU pipeline by testing our best-performing NLU models on noisy ASR output to inspect the challenges of understanding children for math learning in authentic homes.
comment: Proceedings of the 18th Workshop on Innovative Use of NLP for Building Educational Applications (BEA) at ACL 2023
☆ Parallel Neurosymbolic Integration with Concordia
Parallel neurosymbolic architectures have been applied effectively in NLP by distilling knowledge from a logic theory into a deep model.However, prior art faces several limitations including supporting restricted forms of logic theories and relying on the assumption of independence between the logic and the deep network. We present Concordia, a framework overcoming the limitations of prior art. Concordia is agnostic both to the deep network and the logic theory offering support for a wide range of probabilistic theories. Our framework can support supervised training of both components and unsupervised training of the neural component. Concordia has been successfully applied to tasks beyond NLP and data classification, improving the accuracy of state-of-the-art on collective activity detection, entity linking and recommendation tasks.
comment: Fortieth International Conference on Machine Learning, 16 pages (including appendix)
☆ Make Your Pre-trained Model Reversible: From Parameter to Memory Efficient Fine-Tuning
Parameter-efficient fine-tuning (PEFT) of pre-trained language models (PLMs) has emerged as a highly successful approach, with training only a small number of parameters without sacrificing performance and becoming the de-facto learning paradigm with the increasing size of PLMs. However, existing PEFT methods are not memory-efficient, because they still require caching most of the intermediate activations for the gradient calculation, akin to fine-tuning. One effective way to reduce the activation memory is to apply a reversible model, so the intermediate activations are not necessary to be cached and can be recomputed. Nevertheless, modifying a PLM to its reversible variant with PEFT is not straightforward, since the reversible model has a distinct architecture from the currently released PLMs. In this paper, we first investigate what is a key factor for the success of existing PEFT methods, and realize that it's essential to preserve the PLM's starting point when initializing a PEFT method. With this finding, we propose memory-efficient fine-tuning (MEFT) that inserts adapters into a PLM, preserving the PLM's starting point and making it reversible without additional pre-training. We evaluate MEFT on the GLUE benchmark and five question-answering tasks with various backbones, BERT, RoBERTa, BART and OPT. MEFT significantly reduces the activation memory up to 84% of full fine-tuning with a negligible amount of trainable parameters. Moreover, MEFT achieves the same score on GLUE and a comparable score on the question-answering tasks as full fine-tuning.
comment: Code at https://github.com/BaohaoLiao/mefts
☆ Exploring Anisotropy and Outliers in Multilingual Language Models for Cross-Lingual Semantic Sentence Similarity ACL
Previous work has shown that the representations output by contextual language models are more anisotropic than static type embeddings, and typically display outlier dimensions. This seems to be true for both monolingual and multilingual models, although much less work has been done on the multilingual context. Why these outliers occur and how they affect the representations is still an active area of research. We investigate outlier dimensions and their relationship to anisotropy in multiple pre-trained multilingual language models. We focus on cross-lingual semantic similarity tasks, as these are natural tasks for evaluating multilingual representations. Specifically, we examine sentence representations. Sentence transformers which are fine-tuned on parallel resources (that are not always available) perform better on this task, and we show that their representations are more isotropic. However, we aim to improve multilingual representations in general. We investigate how much of the performance difference can be made up by only transforming the embedding space without fine-tuning, and visualise the resulting spaces. We test different operations: Removing individual outlier dimensions, cluster-based isotropy enhancement, and ZCA whitening. We publish our code for reproducibility.
comment: To appear in ACL Findings 2023
☆ A big data approach towards sarcasm detection in Russian
We present a set of deterministic algorithms for Russian inflection and automated text synthesis. These algorithms are implemented in a publicly available web-service www.passare.ru. This service provides functions for inflection of single words, word matching and synthesis of grammatically correct Russian text. Selected code and datasets are available at https://github.com/passare-ru/PassareFunctions/ Performance of the inflectional functions has been tested against the annotated corpus of Russian language OpenCorpora, compared with that of other solutions, and used for estimating the morphological variability and complexity of different parts of speech in Russian.
comment: arXiv admin note: substantial text overlap with arXiv:1706.02551
☆ Responsibility Perspective Transfer for Italian Femicide News ACL 2023
Different ways of linguistically expressing the same real-world event can lead to different perceptions of what happened. Previous work has shown that different descriptions of gender-based violence (GBV) influence the reader's perception of who is to blame for the violence, possibly reinforcing stereotypes which see the victim as partly responsible, too. As a contribution to raise awareness on perspective-based writing, and to facilitate access to alternative perspectives, we introduce the novel task of automatically rewriting GBV descriptions as a means to alter the perceived level of responsibility on the perpetrator. We present a quasi-parallel dataset of sentences with low and high perceived responsibility levels for the perpetrator, and experiment with unsupervised (mBART-based), zero-shot and few-shot (GPT3-based) methods for rewriting sentences. We evaluate our models using a questionnaire study and a suite of automatic metrics.
comment: Accepted for publication in Findings of ACL 2023
☆ How Many Answers Should I Give? An Empirical Study of Multi-Answer Reading Comprehension ACL 2023
The multi-answer phenomenon, where a question may have multiple answers scattered in the document, can be well handled by humans but is challenging enough for machine reading comprehension (MRC) systems. Despite recent progress in multi-answer MRC, there lacks a systematic analysis of how this phenomenon arises and how to better address it. In this work, we design a taxonomy to categorize commonly-seen multi-answer MRC instances, with which we inspect three multi-answer datasets and analyze where the multi-answer challenge comes from. We further analyze how well different paradigms of current multi-answer MRC models deal with different types of multi-answer instances. We find that some paradigms capture well the key information in the questions while others better model the relationship between questions and contexts. We thus explore strategies to make the best of the strengths of different paradigms. Experiments show that generation models can be a promising platform to incorporate different paradigms. Our annotations and code are released for further research.
comment: Findings of ACL 2023
☆ Divide, Conquer, and Combine: Mixture of Semantic-Independent Experts for Zero-Shot Dialogue State Tracking ACL 2023
Zero-shot transfer learning for Dialogue State Tracking (DST) helps to handle a variety of task-oriented dialogue domains without the cost of collecting in-domain data. Existing works mainly study common data- or model-level augmentation methods to enhance the generalization but fail to effectively decouple the semantics of samples, limiting the zero-shot performance of DST. In this paper, we present a simple and effective "divide, conquer and combine" solution, which explicitly disentangles the semantics of seen data, and leverages the performance and robustness with the mixture-of-experts mechanism. Specifically, we divide the seen data into semantically independent subsets and train corresponding experts, the newly unseen samples are mapped and inferred with mixture-of-experts with our designed ensemble inference. Extensive experiments on MultiWOZ2.1 upon the T5-Adapter show our schema significantly and consistently improves the zero-shot performance, achieving the SOTA on settings without external knowledge, with only 10M trainable parameters1.
comment: Accepted to ACL 2023
☆ End-to-end Knowledge Retrieval with Multi-modal Queries ACL 2023
We investigate knowledge retrieval with multi-modal queries, i.e. queries containing information split across image and text inputs, a challenging task that differs from previous work on cross-modal retrieval. We curate a new dataset called ReMuQ for benchmarking progress on this task. ReMuQ requires a system to retrieve knowledge from a large corpus by integrating contents from both text and image queries. We introduce a retriever model ``ReViz'' that can directly process input text and images to retrieve relevant knowledge in an end-to-end fashion without being dependent on intermediate modules such as object detectors or caption generators. We introduce a new pretraining task that is effective for learning knowledge retrieval with multimodal queries and also improves performance on downstream tasks. We demonstrate superior performance in retrieval on two datasets (ReMuQ and OK-VQA) under zero-shot settings as well as further improvements when finetuned on these datasets.
comment: ACL 2023
☆ Uncertainty-Aware Unlikelihood Learning Improves Generative Aspect Sentiment Quad Prediction
Recently, aspect sentiment quad prediction has received widespread attention in the field of aspect-based sentiment analysis. Existing studies extract quadruplets via pre-trained generative language models to paraphrase the original sentence into a templated target sequence. However, previous works only focus on what to generate but ignore what not to generate. We argue that considering the negative samples also leads to potential benefits. In this work, we propose a template-agnostic method to control the token-level generation, which boosts original learning and reduces mistakes simultaneously. Specifically, we introduce Monte Carlo dropout to understand the built-in uncertainty of pre-trained language models, acquiring the noises and errors. We further propose marginalized unlikelihood learning to suppress the uncertainty-aware mistake tokens. Finally, we introduce minimization entropy to balance the effects of marginalized unlikelihood learning. Extensive experiments on four public datasets demonstrate the effectiveness of our approach on various generation templates1.
☆ Towards hate speech detection in low-resource languages: Comparing ASR to acoustic word embeddings on Wolof and Swahili
We consider hate speech detection through keyword spotting on radio broadcasts. One approach is to build an automatic speech recognition (ASR) system for the target low-resource language. We compare this to using acoustic word embedding (AWE) models that map speech segments to a space where matching words have similar vectors. We specifically use a multilingual AWE model trained on labelled data from well-resourced languages to spot keywords in data in the unseen target language. In contrast to ASR, the AWE approach only requires a few keyword exemplars. In controlled experiments on Wolof and Swahili where training and test data are from the same domain, an ASR model trained on just five minutes of data outperforms the AWE approach. But in an in-the-wild test on Swahili radio broadcasts with actual hate speech keywords, the AWE model (using one minute of template data) is more robust, giving similar performance to an ASR system trained on 30 hours of labelled data.
comment: Accepted to Interspeech 2023
☆ BiSync: A Bilingual Editor for Synchronized Monolingual Texts ACL 2023
In our globalized world, a growing number of situations arise where people are required to communicate in one or several foreign languages. In the case of written communication, users with a good command of a foreign language may find assistance from computer-aided translation (CAT) technologies. These technologies often allow users to access external resources, such as dictionaries, terminologies or bilingual concordancers, thereby interrupting and considerably hindering the writing process. In addition, CAT systems assume that the source sentence is fixed and also restrict the possible changes on the target side. In order to make the writing process smoother, we present BiSync, a bilingual writing assistant that allows users to freely compose text in two languages, while maintaining the two monolingual texts synchronized. We also include additional functionalities, such as the display of alternative prefix translations and paraphrases, which are intended to facilitate the authoring of texts. We detail the model architecture used for synchronization and evaluate the resulting tool, showing that high accuracy can be attained with limited computational resources. The interface and models are publicly available at https://github.com/jmcrego/BiSync and a demonstration video can be watched on YouTube at https://youtu.be/_l-ugDHfNgU .
comment: ACL 2023 System Demo
☆ Preference-grounded Token-level Guidance for Language Model Fine-tuning
Aligning language models (LMs) with preferences is an important problem in natural language generation. A key challenge is that preferences are typically provided at the sequence level while LM training and generation both occur at the token level. There is, therefore, a granularity mismatch between the preference and the LM training losses, which may complicate the learning problem. In this paper, we address this issue by developing an alternate training process, where we iterate between grounding the sequence-level preference into token-level training guidance, and improving the LM with the learned guidance. For guidance learning, we design a framework that extends the pairwise-preference learning in imitation learning to both variable-length LM generation and utilizing the preference among multiple generations. For LM training, based on the amount of supervised data, we present two minimalist learning objectives that utilize the learned guidance. In experiments, our method performs competitively on two distinct representative LM tasks -- discrete-prompt generation and text summarization.
☆ Developing and Building Ontologies in Cyber Security
Cyber Security is one of the most arising disciplines in our modern society. We work on Cybersecurity domain and in this the topic we chose is Cyber Security Ontologies. In this we gather all latest and previous ontologies and compare them on the basis of different analyzing factors to get best of them. Reason to select this topic is to assemble different ontologies from different era of time. Because, researches that included in this SLR is mostly studied single ontology. If any researcher wants to study ontologies, he has to study every single ontology and select which one is best for his research. So, we assemble different types of ontology and compare them against each other to get best of them. A total 24 papers between years 2010-2020 are carefully selected through systematic process and classified accordingly. Lastly, this SLR have been presented to provide the researchers promising future directions in the domain of cybersecurity ontologies.
comment: 8 pages, 2 figures
☆ CFL: Causally Fair Language Models Through Token-level Attribute Controlled Generation ACL 2023
We propose a method to control the attributes of Language Models (LMs) for the text generation task using Causal Average Treatment Effect (ATE) scores and counterfactual augmentation. We explore this method, in the context of LM detoxification, and propose the Causally Fair Language (CFL) architecture for detoxifying pre-trained LMs in a plug-and-play manner. Our architecture is based on a Structural Causal Model (SCM) that is mathematically transparent and computationally efficient as compared with many existing detoxification techniques. We also propose several new metrics that aim to better understand the behaviour of LMs in the context of toxic text generation. Further, we achieve state of the art performance for toxic degeneration, which are computed using \RTP (RTP) benchmark. Our experiments show that CFL achieves such a detoxification without much impact on the model perplexity. We also show that CFL mitigates the unintended bias problem through experiments on the BOLD dataset.
comment: 19 pages, 10 figures. Findings of ACL 2023
☆ Focused Prefix Tuning for Controllable Text Generation ACL 2023
In a controllable text generation dataset, there exist unannotated attributes that could provide irrelevant learning signals to models that use it for training and thus degrade their performance. We propose focused prefix tuning(FPT) to mitigate the problem and to enable the control to focus on the desired attribute. Experimental results show that FPT can achieve better control accuracy and text fluency than baseline models in single-attribute control tasks. In multi-attribute control tasks, FPT achieves comparable control accuracy with the state-of-the-art approach while keeping the flexibility to control new attributes without retraining existing models.
comment: Accepted to the ACL 2023
☆ CAISA at SemEval-2023 Task 8: Counterfactual Data Augmentation for Mitigating Class Imbalance in Causal Claim Identification
The class imbalance problem can cause machine learning models to produce an undesirable performance on the minority class as well as the whole dataset. Using data augmentation techniques to increase the number of samples is one way to tackle this problem. We introduce a novel counterfactual data augmentation by verb replacement for the identification of medical claims. In addition, we investigate the impact of this method and compare it with 3 other data augmentation techniques, showing that the proposed method can result in a significant (relative) improvement in the minority class.
☆ CapText: Large Language Model-based Caption Generation From Image Context and Description
While deep-learning models have been shown to perform well on image-to-text datasets, it is difficult to use them in practice for captioning images. This is because \textit{captions} traditionally tend to be context-dependent and offer complementary information about an image, while models tend to produce \textit{descriptions} that describe the visual features of the image. Prior research in caption generation has explored the use of models that generate captions when provided with the images alongside their respective descriptions or contexts. We propose and evaluate a new approach, which leverages existing large language models to generate captions from textual descriptions and context alone, without ever processing the image directly. We demonstrate that after fine-tuning, our approach outperforms current state-of-the-art image-text alignment models like OSCAR-VinVL on this task on the CIDEr metric.
☆ Training-free Neural Architecture Search for RNNs and Transformers
Neural architecture search (NAS) has allowed for the automatic creation of new and effective neural network architectures, offering an alternative to the laborious process of manually designing complex architectures. However, traditional NAS algorithms are slow and require immense amounts of computing power. Recent research has investigated training-free NAS metrics for image classification architectures, drastically speeding up search algorithms. In this paper, we investigate training-free NAS metrics for recurrent neural network (RNN) and BERT-based transformer architectures, targeted towards language modeling tasks. First, we develop a new training-free metric, named hidden covariance, that predicts the trained performance of an RNN architecture and significantly outperforms existing training-free metrics. We experimentally evaluate the effectiveness of the hidden covariance metric on the NAS-Bench-NLP benchmark. Second, we find that the current search space paradigm for transformer architectures is not optimized for training-free neural architecture search. Instead, a simple qualitative analysis can effectively shrink the search space to the best performing architectures. This conclusion is based on our investigation of existing training-free metrics and new metrics developed from recent transformer pruning literature, evaluated on our own benchmark of trained BERT architectures. Ultimately, our analysis shows that the architecture search space and the training-free metric must be developed together in order to achieve effective results.
comment: Code is available at https://github.com/aaronserianni/training-free-nas
☆ AfriNames: Most ASR models "butcher" African Names
Useful conversational agents must accurately capture named entities to minimize error for downstream tasks, for example, asking a voice assistant to play a track from a certain artist, initiating navigation to a specific location, or documenting a laboratory result for a patient. However, where named entities such as ``Ukachukwu`` (Igbo), ``Lakicia`` (Swahili), or ``Ingabire`` (Rwandan) are spoken, automatic speech recognition (ASR) models' performance degrades significantly, propagating errors to downstream systems. We model this problem as a distribution shift and demonstrate that such model bias can be mitigated through multilingual pre-training, intelligent data augmentation strategies to increase the representation of African-named entities, and fine-tuning multilingual ASR models on multiple African accents. The resulting fine-tuned models show an 81.5\% relative WER improvement compared with the baseline on samples with African-named entities.
comment: Accepted at Interspeech 2023 (Main Conference)
☆ Adapting an Unadaptable ASR System INTERSPEECH
As speech recognition model sizes and training data requirements grow, it is increasingly common for systems to only be available via APIs from online service providers rather than having direct access to models themselves. In this scenario it is challenging to adapt systems to a specific target domain. To address this problem we consider the recently released OpenAI Whisper ASR as an example of a large-scale ASR system to assess adaptation methods. An error correction based approach is adopted, as this does not require access to the model, but can be trained from either 1-best or N-best outputs that are normally available via the ASR API. LibriSpeech is used as the primary target domain for adaptation. The generalization ability of the system in two distinct dimensions are then evaluated. First, whether the form of correction model is portable to other speech recognition domains, and secondly whether it can be used for ASR models having a different architecture.
comment: submitted to INTERSPEECH
☆ Estimating Semantic Similarity between In-Domain and Out-of-Domain Samples
Prior work typically describes out-of-domain (OOD) or out-of-distribution (OODist) samples as those that originate from dataset(s) or source(s) different from the training set but for the same task. When compared to in-domain (ID) samples, the models have been known to usually perform poorer on OOD samples, although this observation is not consistent. Another thread of research has focused on OOD detection, albeit mostly using supervised approaches. In this work, we first consolidate and present a systematic analysis of multiple definitions of OOD and OODist as discussed in prior literature. Then, we analyze the performance of a model under ID and OOD/OODist settings in a principled way. Finally, we seek to identify an unsupervised method for reliably identifying OOD/OODist samples without using a trained model. The results of our extensive evaluation using 12 datasets from 4 different tasks suggest the promising potential of unsupervised metrics in this task.
☆ Learning When to Speak: Latency and Quality Trade-offs for Simultaneous Speech-to-Speech Translation with Offline Models INTERSPEECH 2023
Recent work in speech-to-speech translation (S2ST) has focused primarily on offline settings, where the full input utterance is available before any output is given. This, however, is not reasonable in many real-world scenarios. In latency-sensitive applications, rather than waiting for the full utterance, translations should be spoken as soon as the information in the input is present. In this work, we introduce a system for simultaneous S2ST targeting real-world use cases. Our system supports translation from 57 languages to English with tunable parameters for dynamically adjusting the latency of the output -- including four policies for determining when to speak an output sequence. We show that these policies achieve offline-level accuracy with minimal increases in latency over a Greedy (wait-$k$) baseline. We open-source our evaluation code and interactive test script to aid future SimulS2ST research and application development.
comment: To appear at INTERSPEECH 2023
☆ Multi-Dimensional Evaluation of Text Summarization with In-Context Learning ACL
Evaluation of natural language generation (NLG) is complex and multi-dimensional. Generated text can be evaluated for fluency, coherence, factuality, or any other dimensions of interest. Most frameworks that perform such multi-dimensional evaluation require training on large manually or synthetically generated datasets. In this paper, we study the efficacy of large language models as multi-dimensional evaluators using in-context learning, obviating the need for large training datasets. Our experiments show that in-context learning-based evaluators are competitive with learned evaluation frameworks for the task of text summarization, establishing state-of-the-art on dimensions such as relevance and factual consistency. We then analyze the effects of factors such as the selection and number of in-context examples on performance. Finally, we study the efficacy of in-context learning based evaluators in evaluating zero-shot summaries written by large language models such as GPT-3.
comment: ACL Findings '23
☆ Systematic Evaluation of GPT-3 for Zero-Shot Personality Estimation WASSA
Very large language models (LLMs) perform extremely well on a spectrum of NLP tasks in a zero-shot setting. However, little is known about their performance on human-level NLP problems which rely on understanding psychological concepts, such as assessing personality traits. In this work, we investigate the zero-shot ability of GPT-3 to estimate the Big 5 personality traits from users' social media posts. Through a set of systematic experiments, we find that zero-shot GPT-3 performance is somewhat close to an existing pre-trained SotA for broad classification upon injecting knowledge about the trait in the prompts. However, when prompted to provide fine-grained classification, its performance drops to close to a simple most frequent class (MFC) baseline. We further analyze where GPT-3 performs better, as well as worse, than a pretrained lexical model, illustrating systematic errors that suggest ways to improve LLMs on human-level NLP tasks.
comment: Short Paper (5 pages), Accepted to (WASSA) 13th Workshop on Computational Approaches to Subjectivity, Sentiment & Social Media Analysis at ACL 2023
☆ Hybrid Long Document Summarization using C2F-FAR and ChatGPT: A Practical Study
Text summarization is a downstream natural language processing (NLP) task that challenges the understanding and generation capabilities of language models. Considerable progress has been made in automatically summarizing short texts, such as news articles, often leading to satisfactory results. However, summarizing long documents remains a major challenge. This is due to the complex contextual information in the text and the lack of open-source benchmarking datasets and evaluation frameworks that can be used to develop and test model performance. In this work, we use ChatGPT, the latest breakthrough in the field of large language models (LLMs), together with the extractive summarization model C2F-FAR (Coarse-to-Fine Facet-Aware Ranking) to propose a hybrid extraction and summarization pipeline for long documents such as business articles and books. We work with the world-renowned company getAbstract AG and leverage their expertise and experience in professional book summarization. A practical study has shown that machine-generated summaries can perform at least as well as human-written summaries when evaluated using current automated evaluation metrics. However, a closer examination of the texts generated by ChatGPT through human evaluations has shown that there are still critical issues in terms of text coherence, faithfulness, and style. Overall, our results show that the use of ChatGPT is a very promising but not yet mature approach for summarizing long documents and can at best serve as an inspiration for human editors. We anticipate that our work will inform NLP researchers about the extent to which ChatGPT's capabilities for summarizing long documents overlap with practitioners' needs. Further work is needed to test the proposed hybrid summarization pipeline, in particular involving GPT-4, and to propose a new evaluation framework tailored to the task of summarizing long documents.
☆ Leveraging Natural Language Processing For Public Health Screening On YouTube: A COVID-19 Case Study
Background: Social media platforms have become a viable source of medical information, with patients and healthcare professionals using them to share health-related information and track diseases. Similarly, YouTube, the largest video-sharing platform in the world contains vlogs where individuals talk about their illnesses. The aim of our study was to investigate the use of Natural Language Processing (NLP) to identify the spoken content of YouTube vlogs related to the diagnosis of Coronavirus disease of 2019 (COVID-19) for public health screening. Methods: COVID-19 videos on YouTube were searched using relevant keywords. A total of 1000 videos being spoken in English were downloaded out of which 791 were classified as vlogs, 192 were non-vlogs, and 17 were deleted by the channel. The videos were converted into a textual format using Microsoft Streams. The textual data was preprocessed using basic and advanced preprocessing methods. A lexicon of 200 words was created which contained words related to COVID-19. The data was analyzed using topic modeling, word clouds, and lexicon matching. Results: The word cloud results revealed discussions about COVID-19 symptoms like "fever", along with generic terms such as "mask" and "isolation". Lexical analysis demonstrated that in 96.46% of videos, patients discussed generic terms, and in 95.45% of videos, people talked about COVID-19 symptoms. LDA Topic Modeling results also generated topics that successfully captured key themes and content related to our investigation of COVID-19 diagnoses in YouTube vlogs. Conclusion: By leveraging NLP techniques on YouTube vlogs public health practitioners can enhance their ability to mitigate the effects of pandemics and effectively respond to public health challenges.
☆ Faster Causal Attention Over Large Sequences Through Sparse Flash Attention
Transformer-based language models have found many diverse applications requiring them to process sequences of increasing length. For these applications, the causal self-attention -- which is the only component scaling quadratically w.r.t. the sequence length -- becomes a central concern. While many works have proposed schemes to sparsify the attention patterns and reduce the computational overhead of self-attention, those are often limited by implementations concerns and end up imposing a simple and static structure over the attention matrix. Conversely, implementing more dynamic sparse attentions often results in runtimes significantly slower than computing the full attention using the Flash implementation from Dao et al. (2022). We extend FlashAttention to accommodate a large class of attention sparsity patterns that, in particular, encompass key/query dropping and hashing-based attention. This leads to implementations with no computational complexity overhead and a multi-fold runtime speedup on top of FlashAttention. Even with relatively low degrees of sparsity, our method improves visibly upon FlashAttention as the sequence length increases. Without sacrificing perplexity, we increase the training speed of a transformer language model by $2.0\times$ and $3.3\times$ for sequences of respectively $8k$ and $16k$ tokens.
☆ Diverse and Faithful Knowledge-Grounded Dialogue Generation via Sequential Posterior Inference
The capability to generate responses with diversity and faithfulness using factual knowledge is paramount for creating a human-like, trustworthy dialogue system. Common strategies either adopt a two-step paradigm, which optimizes knowledge selection and response generation separately, and may overlook the inherent correlation between these two tasks, or leverage conditional variational method to jointly optimize knowledge selection and response generation by employing an inference network. In this paper, we present an end-to-end learning framework, termed Sequential Posterior Inference (SPI), capable of selecting knowledge and generating dialogues by approximately sampling from the posterior distribution. Unlike other methods, SPI does not require the inference network or assume a simple geometry of the posterior distribution. This straightforward and intuitive inference procedure of SPI directly queries the response generation model, allowing for accurate knowledge selection and generation of faithful responses. In addition to modeling contributions, our experimental results on two common dialogue datasets (Wizard of Wikipedia and Holl-E) demonstrate that SPI outperforms previous strong baselines according to both automatic and human evaluation metrics.
☆ Did You Read the Instructions? Rethinking the Effectiveness of Task Definitions in Instruction Learning ACL 2023
Large language models (LLMs) have shown impressive performance in following natural language instructions to solve unseen tasks. However, it remains unclear whether models truly understand task definitions and whether the human-written definitions are optimal. In this paper, we systematically study the role of task definitions in instruction learning. We first conduct an ablation analysis informed by human annotations to understand which parts of a task definition are most important, and find that model performance only drops substantially when removing contents describing the task output, in particular label information. Next, we propose an automatic algorithm to compress task definitions to a minimal supporting set of tokens, and find that 60\% of tokens can be removed while maintaining or even improving model performance. Based on these results, we propose two strategies to help models better leverage task instructions: (1) providing only key information for tasks in a common structured format, and (2) adding a meta-tuning stage to help the model better understand the definitions. With these two strategies, we achieve a 4.2 Rouge-L improvement over 119 unseen test tasks.
comment: ACL 2023, camera-ready; 10 pages
☆ Evaluating the Capabilities of Multi-modal Reasoning Models with Synthetic Task Data
The impressive advances and applications of large language and joint language-and-visual understanding models has led to an increased need for methods of probing their potential reasoning capabilities. However, the difficulty of gather naturally-occurring data for complex multi-modal reasoning tasks bottlenecks the evaluation of AI methods on tasks which are not already covered by an academic dataset. In this work, we leverage recent advances in high resolution text-to-image generation to develop a framework for generating evaluation data for multi-modal reasoning tasks. We apply this framework to generate context-dependent anomaly data, creating a synthetic dataset on a challenging task which is not well covered by existing datasets. We benchmark the performance of a state-of-the-art visual question answering (VQA) model against data generated with this method, and demonstrate that while the task is tractable, the model performs significantly worse on the context-dependent anomaly detection task than on standard VQA tasks.
☆ Learning Transformer Programs
Recent research in mechanistic interpretability has attempted to reverse-engineer Transformer models by carefully inspecting network weights and activations. However, these approaches require considerable manual effort and still fall short of providing complete, faithful descriptions of the underlying algorithms. In this work, we introduce a procedure for training Transformers that are mechanistically interpretable by design. We build on RASP [Weiss et al., 2021], a programming language that can be compiled into Transformer weights. Instead of compiling human-written programs into Transformers, we design a modified Transformer that can be trained using gradient-based optimization and then be automatically converted into a discrete, human-readable program. We refer to these models as Transformer Programs. To validate our approach, we learn Transformer Programs for a variety of problems, including an in-context learning task, a suite of algorithmic problems (e.g. sorting, recognizing Dyck-languages), and NLP tasks including named entity recognition and text classification. The Transformer Programs can automatically find reasonable solutions, performing on par with standard Transformers of comparable size; and, more importantly, they are easy to interpret. To demonstrate these advantages, we convert Transformers into Python programs and use off-the-shelf code analysis tools to debug model errors and identify the ``circuits'' used to solve different sub-problems. We hope that Transformer Programs open a new path toward the goal of intrinsically interpretable machine learning.
comment: Our code, and example Transformer Programs, are available at https://github.com/princeton-nlp/TransformerPrograms
☆ Examining the Causal Effect of First Names on Language Models: The Case of Social Commonsense Reasoning
As language models continue to be integrated into applications of personal and societal relevance, ensuring these models' trustworthiness is crucial, particularly with respect to producing consistent outputs regardless of sensitive attributes. Given that first names may serve as proxies for (intersectional) socio-demographic representations, it is imperative to examine the impact of first names on commonsense reasoning capabilities. In this paper, we study whether a model's reasoning given a specific input differs based on the first names provided. Our underlying assumption is that the reasoning about Alice should not differ from the reasoning about James. We propose and implement a controlled experimental framework to measure the causal effect of first names on commonsense reasoning, enabling us to distinguish between model predictions due to chance and caused by actual factors of interest. Our results indicate that the frequency of first names has a direct effect on model prediction, with less frequent names yielding divergent predictions compared to more frequent names. To gain insights into the internal mechanisms of models that are contributing to these behaviors, we also conduct an in-depth explainable analysis. Overall, our findings suggest that to ensure model robustness, it is essential to augment datasets with more diverse first names during the configuration stage.
☆ The RefinedWeb Dataset for Falcon LLM: Outperforming Curated Corpora with Web Data, and Web Data Only
Large language models are commonly trained on a mixture of filtered web data and curated high-quality corpora, such as social media conversations, books, or technical papers. This curation process is believed to be necessary to produce performant models with broad zero-shot generalization abilities. However, as larger models requiring pretraining on trillions of tokens are considered, it is unclear how scalable is curation and whether we will run out of unique high-quality data soon. At variance with previous beliefs, we show that properly filtered and deduplicated web data alone can lead to powerful models; even significantly outperforming models from the state-of-the-art trained on The Pile. Despite extensive filtering, the high-quality data we extract from the web is still plentiful, and we are able to obtain five trillion tokens from CommonCrawl. We publicly release an extract of 600 billion tokens from our RefinedWeb dataset, and 1.3/7.5B parameters language models trained on it.
☆ Revisiting Hate Speech Benchmarks: From Data Curation to System Deployment KDD'23
Social media is awash with hateful content, much of which is often veiled with linguistic and topical diversity. The benchmark datasets used for hate speech detection do not account for such divagation as they are predominantly compiled using hate lexicons. However, capturing hate signals becomes challenging in neutrally-seeded malicious content. Thus, designing models and datasets that mimic the real-world variability of hate warrants further investigation. To this end, we present GOTHate, a large-scale code-mixed crowdsourced dataset of around 51k posts for hate speech detection from Twitter. GOTHate is neutrally seeded, encompassing different languages and topics. We conduct detailed comparisons of GOTHate with the existing hate speech datasets, highlighting its novelty. We benchmark it with 10 recent baselines. Our extensive empirical and benchmarking experiments suggest that GOTHate is hard to classify in a text-only setup. Thus, we investigate how adding endogenous signals enhances the hate speech detection task. We augment GOTHate with the user's timeline information and ego network, bringing the overall data source closer to the real-world setup for understanding hateful content. Our proposed solution HEN-mBERT is a modular, multilingual, mixture-of-experts model that enriches the linguistic subspace with latent endogenous signals from history, topology, and exemplars. HEN-mBERT transcends the best baseline by 2.5% and 5% in overall macro-F1 and hate class F1, respectively. Inspired by our experiments, in partnership with Wipro AI, we are developing a semi-automated pipeline to detect hateful content as a part of their mission to tackle online harm.
comment: 15 pages, 4 figures, 11 tables. Accepted at SIGKDD'23
☆ LLMatic: Neural Architecture Search via Large Language Models and Quality-Diversity Optimization
Large Language Models (LLMs) have emerged as powerful tools capable of accomplishing a broad spectrum of tasks. Their abilities span numerous areas, and one area where they have made a significant impact is in the domain of code generation. In this context, we view LLMs as mutation and crossover tools. Meanwhile, Quality-Diversity (QD) algorithms are known to discover diverse and robust solutions. By merging the code-generating abilities of LLMs with the diversity and robustness of QD solutions, we introduce LLMatic, a Neural Architecture Search (NAS) algorithm. While LLMs struggle to conduct NAS directly through prompts, LLMatic uses a procedural approach, leveraging QD for prompts and network architecture to create diverse and highly performant networks. We test LLMatic on the CIFAR-10 image classification benchmark, demonstrating that it can produce competitive networks with just $2,000$ searches, even without prior knowledge of the benchmark domain or exposure to any previous top-performing models for the benchmark.
☆ UCAS-IIE-NLP at SemEval-2023 Task 12: Enhancing Generalization of Multilingual BERT for Low-resource Sentiment Analysis SemEval
This paper describes our system designed for SemEval-2023 Task 12: Sentiment analysis for African languages. The challenge faced by this task is the scarcity of labeled data and linguistic resources in low-resource settings. To alleviate these, we propose a generalized multilingual system SACL-XLMR for sentiment analysis on low-resource languages. Specifically, we design a lexicon-based multilingual BERT to facilitate language adaptation and sentiment-aware representation learning. Besides, we apply a supervised adversarial contrastive learning technique to learn sentiment-spread structured representations and enhance model generalization. Our system achieved competitive results, largely outperforming baselines on both multilingual and zero-shot sentiment classification subtasks. Notably, the system obtained the 1st rank on the zero-shot classification subtask in the official ranking. Extensive experiments demonstrate the effectiveness of our system.
comment: 9 pages, accepted by SemEval@ACL 2023
☆ Improving the Robustness of Summarization Systems with Dual Augmentation ACL 2023
A robust summarization system should be able to capture the gist of the document, regardless of the specific word choices or noise in the input. In this work, we first explore the summarization models' robustness against perturbations including word-level synonym substitution and noise. To create semantic-consistent substitutes, we propose a SummAttacker, which is an efficient approach to generating adversarial samples based on language models. Experimental results show that state-of-the-art summarization models have a significant decrease in performance on adversarial and noisy test sets. Next, we analyze the vulnerability of the summarization systems and explore improving the robustness by data augmentation. Specifically, the first brittleness factor we found is the poor understanding of infrequent words in the input. Correspondingly, we feed the encoder with more diverse cases created by SummAttacker in the input space. The other factor is in the latent space, where the attacked inputs bring more variations to the hidden states. Hence, we construct adversarial decoder input and devise manifold softmixing operation in hidden space to introduce more diversity. Experimental results on Gigaword and CNN/DM datasets demonstrate that our approach achieves significant improvements over strong baselines and exhibits higher robustness on noisy, attacked, and clean datasets.
comment: 10 pages, 6 figures, ACL 2023 main coference
☆ Quantization-Aware and Tensor-Compressed Training of Transformers for Natural Language Understanding
Fine-tuned transformer models have shown superior performances in many natural language tasks. However, the large model size prohibits deploying high-performance transformer models on resource-constrained devices. This paper proposes a quantization-aware tensor-compressed training approach to reduce the model size, arithmetic operations, and ultimately runtime latency of transformer-based models. We compress the embedding and linear layers of transformers into small low-rank tensor cores, which significantly reduces model parameters. A quantization-aware training with learnable scale factors is used to further obtain low-precision representations of the tensor-compressed models. The developed approach can be used for both end-to-end training and distillation-based training. To improve the convergence, a layer-by-layer distillation is applied to distill a quantized and tensor-compressed student model from a pre-trained transformer. The performance is demonstrated in two natural language understanding tasks, showing up to $63\times$ compression ratio, little accuracy loss and remarkable inference and training speedup.
☆ TimelineQA: A Benchmark for Question Answering over Timelines
Lifelogs are descriptions of experiences that a person had during their life. Lifelogs are created by fusing data from the multitude of digital services, such as online photos, maps, shopping and content streaming services. Question answering over lifelogs can offer personal assistants a critical resource when they try to provide advice in context. However, obtaining answers to questions over lifelogs is beyond the current state of the art of question answering techniques for a variety of reasons, the most pronounced of which is that lifelogs combine free text with some degree of structure such as temporal and geographical information. We create and publicly release TimelineQA1, a benchmark for accelerating progress on querying lifelogs. TimelineQA generates lifelogs of imaginary people. The episodes in the lifelog range from major life episodes such as high school graduation to those that occur on a daily basis such as going for a run. We describe a set of experiments on TimelineQA with several state-of-the-art QA models. Our experiments reveal that for atomic queries, an extractive QA system significantly out-performs a state-of-the-art retrieval-augmented QA system. For multi-hop queries involving aggregates, we show that the best result is obtained with a state-of-the-art table QA technique, assuming the ground truth set of episodes for deriving the answer is available.
☆ Reimagining Retrieval Augmented Language Models for Answering Queries
We present a reality check on large language models and inspect the promise of retrieval augmented language models in comparison. Such language models are semi-parametric, where models integrate model parameters and knowledge from external data sources to make their predictions, as opposed to the parametric nature of vanilla large language models. We give initial experimental findings that semi-parametric architectures can be enhanced with views, a query analyzer/planner, and provenance to make a significantly more powerful system for question answering in terms of accuracy and efficiency, and potentially for other NLP tasks
☆ Are Layout-Infused Language Models Robust to Layout Distribution Shifts? A Case Study with Scientific Documents ACL
Recent work has shown that infusing layout features into language models (LMs) improves processing of visually-rich documents such as scientific papers. Layout-infused LMs are often evaluated on documents with familiar layout features (e.g., papers from the same publisher), but in practice models encounter documents with unfamiliar distributions of layout features, such as new combinations of text sizes and styles, or new spatial configurations of textual elements. In this work we test whether layout-infused LMs are robust to layout distribution shifts. As a case study we use the task of scientific document structure recovery, segmenting a scientific paper into its structural categories (e.g., "title", "caption", "reference"). To emulate distribution shifts that occur in practice we re-partition the GROTOAP2 dataset. We find that under layout distribution shifts model performance degrades by up to 20 F1. Simple training strategies, such as increasing training diversity, can reduce this degradation by over 35% relative F1; however, models fail to reach in-distribution performance in any tested out-of-distribution conditions. This work highlights the need to consider layout distribution shifts during model evaluation, and presents a methodology for conducting such evaluations.
comment: To appear in ACL Findings 2023
☆ Bypass Temporal Classification: Weakly Supervised Automatic Speech Recognition with Imperfect Transcripts
This paper presents a novel algorithm for building an automatic speech recognition (ASR) model with imperfect training data. Imperfectly transcribed speech is a prevalent issue in human-annotated speech corpora, which degrades the performance of ASR models. To address this problem, we propose Bypass Temporal Classification (BTC) as an expansion of the Connectionist Temporal Classification (CTC) criterion. BTC explicitly encodes the uncertainties associated with transcripts during training. This is accomplished by enhancing the flexibility of the training graph, which is implemented as a weighted finite-state transducer (WFST) composition. The proposed algorithm improves the robustness and accuracy of ASR systems, particularly when working with imprecisely transcribed speech corpora. Our implementation will be open-sourced.
♻ ☆ LFTK: Handcrafted Features in Computational Linguistics ACL 2023
Past research has identified a rich set of handcrafted linguistic features that can potentially assist various tasks. However, their extensive number makes it difficult to effectively select and utilize existing handcrafted features. Coupled with the problem of inconsistent implementation across research works, there has been no categorization scheme or generally-accepted feature names. This creates unwanted confusion. Also, most existing handcrafted feature extraction libraries are not open-source or not actively maintained. As a result, a researcher often has to build such an extraction system from the ground up. We collect and categorize more than 220 popular handcrafted features grounded on past literature. Then, we conduct a correlation analysis study on several task-specific datasets and report the potential use cases of each feature. Lastly, we devise a multilingual handcrafted linguistic feature extraction system in a systematically expandable manner. We open-source our system for public access to a rich set of pre-implemented handcrafted features. Our system is coined LFTK and is the largest of its kind. Find it at github.com/brucewlee/lftk.
comment: BEA @ ACL 2023
♻ ☆ Towards Revealing the Mystery behind Chain of Thought: a Theoretical Perspective
Recent studies have discovered that Chain-of-Thought prompting (CoT) can dramatically improve the performance of Large Language Models (LLMs), particularly when dealing with complex tasks involving mathematics or reasoning. Despite the enormous empirical success, the underlying mechanisms behind CoT and how it unlocks the potential of LLMs remain elusive. In this paper, we take a first step towards theoretically answering these questions. Specifically, we examine the expressivity of LLMs with CoT in solving fundamental mathematical and decision-making problems. We start by giving an impossibility result showing that bounded-depth Transformers are unable to directly produce correct answers for basic arithmetic/equation tasks unless the model size grows super-polynomially with respect to the input length. In contrast, we then prove by construction that autoregressive Transformers of constant size suffice to solve both tasks by generating CoT derivations using a commonly-used math language format. Moreover, we show LLMs with CoT are capable of solving a general class of decision-making problems known as Dynamic Programming, thus justifying its power in tackling complex real-world tasks. Finally, extensive experiments on four tasks show that, while Transformers always fail to predict the answers directly, they can consistently learn to generate correct solutions step-by-step given sufficient CoT demonstrations.
comment: 34 pages
♻ ☆ DISCO: Distilling Phrasal Counterfactuals with Large Language Models ACL 2023
Models trained with counterfactually augmented data learn representations of the causal structure of tasks, enabling robust generalization. However, high-quality counterfactual data is scarce for most tasks and not easily generated at scale. When crowdsourced, such data is typically limited in scale and diversity; when generated using supervised methods, it is computationally expensive to extend to new counterfactual dimensions. In this work, we introduce DISCO (DIStilled COunterfactual Data), a new method for automatically generating high quality counterfactual data at scale. DISCO engineers prompts to generate phrasal perturbations with a large general language model. Then, a task-specific teacher model filters these generations to distill high-quality counterfactual data. While task-agnostic, we apply our pipeline to the task of natural language inference (NLI) and find that on challenging evaluations such as the NLI stress test, comparatively smaller student models trained with DISCO generated counterfactuals are more robust (6% absolute) and generalize better across distributions (2%) compared to models trained without data augmentation. Furthermore, DISCO augmented models are 10% more consistent between counterfactual pairs on three evaluation sets, demonstrating that DISCO augmentation enables models to more reliably learn causal representations. Our repository is available at: https://github.com/eric11eca/disco
comment: ACL 2023 camera ready
♻ ☆ Heterogeneous Value Evaluation for Large Language Models
The emergent capabilities of Large Language Models (LLMs) have made it crucial to align their values with those of humans. Current methodologies typically attempt alignment with a homogeneous human value and requires human verification, yet lack consensus on the desired aspect and depth of alignment and resulting human biases. In this paper, we propose A2EHV, an Automated Alignment Evaluation with a Heterogeneous Value system that (1) is automated to minimize individual human biases, and (2) allows assessments against various target values to foster heterogeneous agents. Our approach pivots on the concept of value rationality, which represents the ability for agents to execute behaviors that satisfy a target value the most. The quantification of value rationality is facilitated by the Social Value Orientation framework from social psychology, which partitions the value space into four categories to assess social preferences from agents' behaviors. We evaluate the value rationality of eight mainstream LLMs and observe that large models are more inclined to align neutral values compared to those with strong personal values. By examining the behavior of these LLMs, we contribute to a deeper understanding of value alignment within a heterogeneous value system.
comment: Our full prompts are released in the repo: https://github.com/zowiezhang/A2EHV
♻ ☆ Decision-Oriented Dialogue for Human-AI Collaboration
We describe a class of tasks called decision-oriented dialogues, in which AI assistants must collaborate with one or more humans via natural language to help them make complex decisions. We formalize three domains in which users face everyday decisions: (1) choosing an assignment of reviewers to conference papers, (2) planning a multi-step itinerary in a city, and (3) negotiating travel plans for a group of friends. In each of these settings, AI assistants and users have disparate abilities that they must combine to arrive at the best decision: assistants can access and process large amounts of information, while users have preferences and constraints external to the system. For each task, we build a dialogue environment where agents receive a reward based on the quality of the final decision they reach. Using these environments, we collect human-human dialogues with humans playing the role of assistant. To compare how current AI assistants communicate in these settings, we present baselines using large language models in self-play. Finally, we highlight a number of challenges models face in decision-oriented dialogues, ranging from efficient communication to reasoning and optimization, and release our environments as a testbed for future modeling work.
♻ ☆ Neural Natural Language Processing for Long Texts: A Survey of the State-of-the-Art
The adoption of Deep Neural Networks (DNNs) has greatly benefited Natural Language Processing (NLP) during the past decade. However, the demands of long document analysis are quite different from those of shorter texts, while the ever increasing size of documents uploaded on-line renders automated understanding of long texts a critical area of research. This article has two goals: a) it overviews the relevant neural building blocks, thus serving as a short tutorial, and b) it surveys the state-of-the-art in long document NLP, mainly focusing on two central tasks: document classification and document summarization. Sentiment analysis for long texts is also covered, since it is typically treated as a particular case of document classification. Additionally, this article discusses the main challenges, issues and current solutions related to long document NLP. Finally, the relevant, publicly available, annotated datasets are presented, in order to facilitate further research.
comment: 54 pages, 2 figures, 173 citations
♻ ☆ Arithmetic Sampling: Parallel Diverse Decoding for Large Language Models ICML 2023
Decoding methods for large language models often trade-off between diversity of outputs and parallelism of computation. Methods such as beam search and Gumbel top-k sampling can guarantee a different output for each element of the beam, but are not easy to parallelize. Alternatively, methods such as temperature sampling and its modifications (top-k sampling, nucleus sampling, typical decoding, and others), are embarrassingly parallel, but have no guarantees about duplicate samples. We present a framework for sampling according to an arithmetic code book implicitly defined by a large language model, compatible with common sampling variations, with provable beam diversity under certain conditions, as well as being embarrassingly parallel and providing unbiased and consistent expectations from the original model. We demonstrate the effectiveness of our approach on WMT machine translation, more than halving the standard deviation when estimating expected BLEU score reward, and closing the BLEU score gap between independent sampling and beam search by up to 63%.
comment: 17 pages, to appear at ICML 2023
♻ ☆ SMARAGD: Learning SMatch for Accurate and Rapid Approximate Graph Distance
The similarity of graph structures, such as Meaning Representations (MRs), is often assessed via structural matching algorithms, such as Smatch (Cai and Knight, 2013). However, Smatch involves a combinatorial problem that suffers from NP-completeness, making large-scale applications, e.g., graph clustering or search, infeasible. To alleviate this issue, we learn SMARAGD: Semantic Match for Accurate and Rapid Approximate Graph Distance. We show the potential of neural networks to approximate Smatch scores, i) in linear time using a machine translation framework to predict alignments, or ii) in constant time using a Siamese CNN to directly predict Smatch scores. We show that the approximation error can be substantially reduced through data augmentation and graph anonymization.
comment: to appear at 15th International Conference on Computational Semantics (IWCS 2023)
♻ ☆ Grounding Language Models to Images for Multimodal Inputs and Outputs ICML 2023
We propose an efficient method to ground pretrained text-only language models to the visual domain, enabling them to process arbitrarily interleaved image-and-text data, and generate text interleaved with retrieved images. Our method leverages the abilities of language models learnt from large scale text-only pretraining, such as in-context learning and free-form text generation. We keep the language model frozen, and finetune input and output linear layers to enable cross-modality interactions. This allows our model to process arbitrarily interleaved image-and-text inputs, and generate free-form text interleaved with retrieved images. We achieve strong zero-shot performance on grounded tasks such as contextual image retrieval and multimodal dialogue, and showcase compelling interactive abilities. Our approach works with any off-the-shelf language model and paves the way towards an effective, general solution for leveraging pretrained language models in visually grounded settings.
comment: Published in ICML 2023. Project page: https://jykoh.com/fromage
♻ ☆ Best Prompts for Text-to-Image Models and How to Find Them SIGIR '23
Recent progress in generative models, especially in text-guided diffusion models, has enabled the production of aesthetically-pleasing imagery resembling the works of professional human artists. However, one has to carefully compose the textual description, called the prompt, and augment it with a set of clarifying keywords. Since aesthetics are challenging to evaluate computationally, human feedback is needed to determine the optimal prompt formulation and keyword combination. In this paper, we present a human-in-the-loop approach to learning the most useful combination of prompt keywords using a genetic algorithm. We also show how such an approach can improve the aesthetic appeal of images depicting the same descriptions.
comment: 13 pages (6 main pages), 7 figures, 4 tables, accepted at SIGIR '23 Short Paper Track
♻ ☆ MT Metrics Correlate with Human Ratings of Simultaneous Speech Translation
There have been several meta-evaluation studies on the correlation between human ratings and offline machine translation (MT) evaluation metrics such as BLEU, chrF2, BertScore and COMET. These metrics have been used to evaluate simultaneous speech translation (SST) but their correlations with human ratings of SST, which has been recently collected as Continuous Ratings (CR), are unclear. In this paper, we leverage the evaluations of candidate systems submitted to the English-German SST task at IWSLT 2022 and conduct an extensive correlation analysis of CR and the aforementioned metrics. Our study reveals that the offline metrics are well correlated with CR and can be reliably used for evaluating machine translation in simultaneous mode, with some limitations on the test set size. We conclude that given the current quality levels of SST, these metrics can be used as proxies for CR, alleviating the need for large scale human evaluation. Additionally, we observe that correlations of the metrics with translation as a reference is significantly higher than with simultaneous interpreting, and thus we recommend the former for reliable evaluation.
comment: IWSLT 2023
♻ ☆ SQuId: Measuring Speech Naturalness in Many Languages ICASSP 2023
Much of text-to-speech research relies on human evaluation, which incurs heavy costs and slows down the development process. The problem is particularly acute in heavily multilingual applications, where recruiting and polling judges can take weeks. We introduce SQuId (Speech Quality Identification), a multilingual naturalness prediction model trained on over a million ratings and tested in 65 locales-the largest effort of this type to date. The main insight is that training one model on many locales consistently outperforms mono-locale baselines. We present our task, the model, and show that it outperforms a competitive baseline based on w2v-BERT and VoiceMOS by 50.0%. We then demonstrate the effectiveness of cross-locale transfer during fine-tuning and highlight its effect on zero-shot locales, i.e., locales for which there is no fine-tuning data. Through a series of analyses, we highlight the role of non-linguistic effects such as sound artifacts in cross-locale transfer. Finally, we present the effect of our design decision, e.g., model size, pre-training diversity, and language rebalancing with several ablation experiments.
comment: Accepted at ICASSP 2023, with additional material in the appendix
♻ ☆ Let's have a chat! A Conversation with ChatGPT: Technology, Applications, and Limitations
The emergence of an AI-powered chatbot that can generate human-like sentences and write coherent essays has caught the world's attention. This paper discusses the historical overview of chatbots and the technology behind Chat Generative Pre-trained Transformer, better known as ChatGPT. Moreover, potential applications of ChatGPT in various domains, including healthcare, education, and research, are highlighted. Despite promising results, there are several privacy and ethical concerns surrounding ChatGPT. In addition, we highlight some of the important limitations of the current version of ChatGPT. We also ask ChatGPT to provide its point of view and present its responses to several questions we attempt to answer.
comment: This manuscript has been accepted by Artificial Intelligence and Applications (AIA, ISSN: 2811-0854), 2023
♻ ☆ Beam Tree Recursive Cells ICML 2023
We propose Beam Tree Recursive Cell (BT-Cell) - a backpropagation-friendly framework to extend Recursive Neural Networks (RvNNs) with beam search for latent structure induction. We further extend this framework by proposing a relaxation of the hard top-k operators in beam search for better propagation of gradient signals. We evaluate our proposed models in different out-of-distribution splits in both synthetic and realistic data. Our experiments show that BTCell achieves near-perfect performance on several challenging structure-sensitive synthetic tasks like ListOps and logical inference while maintaining comparable performance in realistic data against other RvNN-based models. Additionally, we identify a previously unknown failure case for neural models in generalization to unseen number of arguments in ListOps. The code is available at: https://github.com/JRC1995/BeamTreeRecursiveCells.
comment: Accepted in ICML 2023
♻ ☆ Neural Task Synthesis for Visual Programming
Generative neural models hold great promise in enhancing programming education by synthesizing new content for students. We seek to design neural models that can automatically generate programming tasks for a given specification in the context of visual programming domains. Despite the recent successes of large generative models like GPT-4, our initial results show that these models are ineffective in synthesizing visual programming tasks and struggle with logical and spatial reasoning. We propose a novel neuro-symbolic technique, NeurTaskSyn, that can synthesize programming tasks for a specification given in the form of desired programming concepts exercised by its solution code and constraints on the visual task. NeurTaskSyn has two components: the first component is trained via imitation learning procedure to generate possible solution codes, and the second component is trained via reinforcement learning procedure to guide an underlying symbolic execution engine that generates visual tasks for these codes. We demonstrate the effectiveness of NeurTaskSyn through an extensive empirical evaluation and a qualitative study on reference tasks taken from the Hour of Code: Classic Maze challenge by Code-dot-org and the Intro to Programming with Karel course by CodeHS-dot-com.
♻ ☆ Deep Learning Methods for Extracting Metaphorical Names of Flowers and Plants
The domain of Botany is rich with metaphorical terms. Those terms play an important role in the description and identification of flowers and plants. However, the identification of such terms in discourse is an arduous task. This leads in some cases to committing errors during translation processes and lexicographic tasks. The process is even more challenging when it comes to machine translation, both in the cases of single-word terms and multi-word terms. One of the recent concerns of Natural Language Processing (NLP) applications and Machine Translation (MT) technologies is the automatic identification of metaphor-based words in discourse through Deep Learning (DL). In this study, we seek to fill this gap through the use of thirteen popular transformer based models, as well as ChatGPT, and we show that discriminative models perform better than GPT-3.5 model with our best performer reporting 92.2349% F1 score in metaphoric flower and plant names identification task.
comment: Accepted for SEPLN 2023
♻ ☆ Claim-Dissector: An Interpretable Fact-Checking System with Joint Re-ranking and Veracity Prediction
We present Claim-Dissector: a novel latent variable model for fact-checking and analysis, which given a claim and a set of retrieved evidences jointly learns to identify: (i) the relevant evidences to the given claim, (ii) the veracity of the claim. We propose to disentangle the per-evidence relevance probability and its contribution to the final veracity probability in an interpretable way -- the final veracity probability is proportional to a linear ensemble of per-evidence relevance probabilities. In this way, the individual contributions of evidences towards the final predicted probability can be identified. In per-evidence relevance probability, our model can further distinguish whether each relevant evidence is supporting (S) or refuting (R) the claim. This allows to quantify how much the S/R probability contributes to the final verdict or to detect disagreeing evidence. Despite its interpretable nature, our system achieves results competitive with state-of-the-art on the FEVER dataset, as compared to typical two-stage system pipelines, while using significantly fewer parameters. It also sets new state-of-the-art on FAVIQ and RealFC datasets. Furthermore, our analysis shows that our model can learn fine-grained relevance cues while using coarse-grained supervision, and we demonstrate it in 2 ways. (i) We show that our model can achieve competitive sentence recall while using only paragraph-level relevance supervision. (ii) Traversing towards the finest granularity of relevance, we show that our model is capable of identifying relevance at the token level. To do this, we present a new benchmark TLR-FEVER focusing on token-level interpretability -- humans annotate tokens in relevant evidences they considered essential when making their judgment. Then we measure how similar are these annotations to the tokens our model is focusing on.
comment: updated acknowledgement
♻ ☆ Hard Prompts Made Easy: Gradient-Based Discrete Optimization for Prompt Tuning and Discovery
The strength of modern generative models lies in their ability to be controlled through text-based prompts. Typical "hard" prompts are made from interpretable words and tokens, and must be hand-crafted by humans. There are also "soft" prompts, which consist of continuous feature vectors. These can be discovered using powerful optimization methods, but they cannot be easily interpreted, re-used across models, or plugged into a text-based interface. We describe an approach to robustly optimize hard text prompts through efficient gradient-based optimization. Our approach automatically generates hard text-based prompts for both text-to-image and text-to-text applications. In the text-to-image setting, the method creates hard prompts for diffusion models, allowing API users to easily generate, discover, and mix and match image concepts without prior knowledge on how to prompt the model. In the text-to-text setting, we show that hard prompts can be automatically discovered that are effective in tuning LMs for classification.
comment: 15 pages, 12 figures, Code is available at https://github.com/YuxinWenRick/hard-prompts-made-easy
♻ ☆ A Survey on In-context Learning
With the increasing ability of large language models (LLMs), in-context learning (ICL) has become a new paradigm for natural language processing (NLP), where LLMs make predictions only based on contexts augmented with a few examples. It has been a new trend to explore ICL to evaluate and extrapolate the ability of LLMs. In this paper, we aim to survey and summarize the progress and challenges of ICL. We first present a formal definition of ICL and clarify its correlation to related studies. Then, we organize and discuss advanced techniques, including training strategies, demonstration designing strategies, as well as related analysis. Finally, we discuss the challenges of ICL and provide potential directions for further research. We hope that our work can encourage more research on uncovering how ICL works and improving ICL.
comment: Papers collected until 2023/05/22
♻ ☆ Teaching Small Language Models to Reason
Chain of thought prompting successfully improves the reasoning capabilities of large language models, achieving state of the art results on a range of datasets. However, these reasoning capabilities only appear to emerge in models with a size of over 100 billion parameters. In this paper, we explore the transfer of such reasoning capabilities to models with less than 100 billion parameters via knowledge distillation. Specifically, we finetune a student model on the chain of thought outputs generated by a larger teacher model. Our experiments show that the proposed method improves task performance across arithmetic, commonsense and symbolic reasoning datasets. For example, the accuracy of T5 XXL on GSM8K improves from 8.11% to 21.99% when finetuned on PaLM-540B generated chains of thought.
♻ ☆ Automatic Emotion Experiencer Recognition
The most prominent subtask in emotion analysis is emotion classification; to assign a category to a textual unit, for instance a social media post. Many research questions from the social sciences do, however, not only require the detection of the emotion of an author of a post but to understand who is ascribed an emotion in text. This task is tackled by emotion role labeling which aims at extracting who is described in text to experience an emotion, why, and towards whom. This could, however, be considered overly sophisticated if the main question to answer is who feels which emotion. A targeted approach for such setup is to classify emotion experiencer mentions (aka "emoters") regarding the emotion they presumably perceive. This task is similar to named entity recognition of person names with the difference that not every mentioned entity name is an emoter. While, very recently, data with emoter annotations has been made available, no experiments have yet been performed to detect such mentions. With this paper, we provide baseline experiments to understand how challenging the task is. We further evaluate the impact on experiencer-specific emotion categorization and appraisal detection in a pipeline, when gold mentions are not available. We show that experiencer detection in text is a challenging task, with a precision of .82 and a recall of .56 (F1 =.66). These results motivate future work of jointly modeling emoter spans and emotion/appraisal predictions.
♻ ☆ DC CoMix TTS: An End-to-End Expressive TTS with Discrete Code Collaborated with Mixer
Despite the huge successes made in neutral TTS, content-leakage remains a challenge. In this paper, we propose a new input representation and simple architecture to achieve improved prosody modeling. Inspired by the recent success in the use of discrete code in TTS, we introduce discrete code to the input of the reference encoder. Specifically, we leverage the vector quantizer from the audio compression model to exploit the diverse acoustic information it has already been trained on. In addition, we apply the modified MLP-Mixer to the reference encoder, making the architecture lighter. As a result, we train the prosody transfer TTS in an end-to-end manner. We prove the effectiveness of our method through both subjective and objective evaluations. We demonstrate that the reference encoder learns better speaker-independent prosody when discrete code is utilized as input in the experiments. In addition, we obtain comparable results even when fewer parameters are inputted.
comment: Accepted at Interspeech 2023
♻ ☆ Bag of Tricks for Training Data Extraction from Language Models ICML 2023
With the advance of language models, privacy protection is receiving more attention. Training data extraction is therefore of great importance, as it can serve as a potential tool to assess privacy leakage. However, due to the difficulty of this task, most of the existing methods are proof-of-concept and still not effective enough. In this paper, we investigate and benchmark tricks for improving training data extraction using a publicly available dataset. Because most existing extraction methods use a pipeline of generating-then-ranking, i.e., generating text candidates as potential training data and then ranking them based on specific criteria, our research focuses on the tricks for both text generation (e.g., sampling strategy) and text ranking (e.g., token-level criteria). The experimental results show that several previously overlooked tricks can be crucial to the success of training data extraction. Based on the GPT-Neo 1.3B evaluation results, our proposed tricks outperform the baseline by a large margin in most cases, providing a much stronger baseline for future research. The code is available at https://github.com/weichen-yu/LM-Extraction.
comment: ICML 2023
♻ ☆ Conformal Prediction with Large Language Models for Multi-Choice Question Answering
As large language models continue to be widely developed, robust uncertainty quantification techniques will become crucial for their safe deployment in high-stakes scenarios. In this work, we explore how conformal prediction can be used to provide uncertainty quantification in language models for the specific task of multiple-choice question-answering. We find that the uncertainty estimates from conformal prediction are tightly correlated with prediction accuracy. This observation can be useful for downstream applications such as selective classification and filtering out low-quality predictions. We also investigate the exchangeability assumption required by conformal prediction to out-of-subject questions, which may be a more realistic scenario for many practical applications. Our work contributes towards more trustworthy and reliable usage of large language models in safety-critical situations, where robust guarantees of error rate are required.
comment: Added additional references
♻ ☆ ChatGPT as a Text Simplification Tool to Remove Bias
The presence of specific linguistic signals particular to a certain sub-group of people can be picked up by language models during training. If the model begins to associate specific language with a distinct group, any decisions made based upon this language would hold a strong correlation to a decision based upon their protected characteristic, leading to possible discrimination. We explore a potential technique for bias mitigation in the form of simplification of text. The driving force of this idea is that simplifying text should standardise language between different sub-groups to one way of speaking while keeping the same meaning. The experiment shows promising results as the classifier accuracy for predicting the sensitive attribute drops by up to 17% for the simplified data.
♻ ☆ Ghost in the Minecraft: Generally Capable Agents for Open-World Environments via Large Language Models with Text-based Knowledge and Memory
The captivating realm of Minecraft has attracted substantial research interest in recent years, serving as a rich platform for developing intelligent agents capable of functioning in open-world environments. However, the current research landscape predominantly focuses on specific objectives, such as the popular "ObtainDiamond" task, and has not yet shown effective generalization to a broader spectrum of tasks. Furthermore, the current leading success rate for the "ObtainDiamond" task stands at around 20%, highlighting the limitations of Reinforcement Learning (RL) based controllers used in existing methods. To tackle these challenges, we introduce Ghost in the Minecraft (GITM), a novel framework integrates Large Language Models (LLMs) with text-based knowledge and memory, aiming to create Generally Capable Agents (GCAs) in Minecraft. These agents, equipped with the logic and common sense capabilities of LLMs, can skillfully navigate complex, sparse-reward environments with text-based interactions. We develop a set of structured actions and leverage LLMs to generate action plans for the agents to execute. The resulting LLM-based agent markedly surpasses previous methods, achieving a remarkable improvement of +47.5% in success rate on the "ObtainDiamond" task, demonstrating superior robustness compared to traditional RL-based controllers. Notably, our agent is the first to procure all items in the Minecraft Overworld technology tree, demonstrating its extensive capabilities. GITM does not need any GPU for training, but a single CPU node with 32 CPU cores is enough. This research shows the potential of LLMs in developing capable agents for handling long-horizon, complex tasks and adapting to uncertainties in open-world environments. See the project website at https://github.com/OpenGVLab/GITM.
♻ ☆ Speaking Multiple Languages Affects the Moral Bias of Language Models ACL
Pre-trained multilingual language models (PMLMs) are commonly used when dealing with data from multiple languages and cross-lingual transfer. However, PMLMs are trained on varying amounts of data for each language. In practice this means their performance is often much better on English than many other languages. We explore to what extent this also applies to moral norms. Do the models capture moral norms from English and impose them on other languages? Do the models exhibit random and thus potentially harmful beliefs in certain languages? Both these issues could negatively impact cross-lingual transfer and potentially lead to harmful outcomes. In this paper, we (1) apply the MoralDirection framework to multilingual models, comparing results in German, Czech, Arabic, Chinese, and English, (2) analyse model behaviour on filtered parallel subtitles corpora, and (3) apply the models to a Moral Foundations Questionnaire, comparing with human responses from different countries. Our experiments demonstrate that, indeed, PMLMs encode differing moral biases, but these do not necessarily correspond to cultural differences or commonalities in human opinions. We release our code and models.
comment: To appear in ACL Findings 2023
♻ ☆ ProsAudit, a prosodic benchmark for self-supervised speech models
We present ProsAudit, a benchmark in English to assess structural prosodic knowledge in self-supervised learning (SSL) speech models. It consists of two subtasks, their corresponding metrics, and an evaluation dataset. In the protosyntax task, the model must correctly identify strong versus weak prosodic boundaries. In the lexical task, the model needs to correctly distinguish between pauses inserted between words and within words. We also provide human evaluation scores on this benchmark. We evaluated a series of SSL models and found that they were all able to perform above chance on both tasks, even when evaluated on an unseen language. However, non-native models performed significantly worse than native ones on the lexical task, highlighting the importance of lexical knowledge in this task. We also found a clear effect of size with models trained on more data performing better in the two subtasks.
comment: Accepted at Interspeech 2023. 4 pages + references, 1 figure
♻ ☆ LMs with a Voice: Spoken Language Modeling beyond Speech Tokens
We present SPECTRON, a novel approach to adapting pre-trained language models (LMs) to perform speech continuation. By leveraging pre-trained speech encoders, our model generates both text and speech outputs with the entire system being trained end-to-end operating directly on spectrograms. Training the entire model in the spectrogram domain simplifies our speech continuation system versus existing cascade methods which use discrete speech representations. We further show our method surpasses existing spoken language models both in semantic content and speaker preservation while also benefiting from the knowledge transferred from pre-existing models. Audio samples can be found in our website https://michelleramanovich.github.io/spectron/spectron
♻ ☆ Translatotron 3: Speech to Speech Translation with Monolingual Data
This paper presents Translatotron 3, a novel approach to train a direct speech-to-speech translation model from monolingual speech-text datasets only in a fully unsupervised manner. Translatotron 3 combines masked autoencoder, unsupervised embedding mapping, and back-translation to achieve this goal. Experimental results in speech-to-speech translation tasks between Spanish and English show that Translatotron 3 outperforms a baseline cascade system, reporting 18.14 BLEU points improvement on the synthesized Unpaired-Conversational dataset. In contrast to supervised approaches that necessitate real paired data, which is unavailable, or specialized modeling to replicate para-/non-linguistic information, Translatotron 3 showcases its capability to retain para-/non-linguistic such as pauses, speaking rates, and speaker identity. Audio samples can be found in our website http://google-research.github.io/lingvo-lab/translatotron3
♻ ☆ BeamSearchQA: Large Language Models are Strong Zero-Shot QA Solver
Open-domain question answering is a crucial task that often requires accessing external information. Existing methods typically adopt a single-turn retrieve-then-read approach, where relevant documents are first retrieved, and questions are then answered based on the retrieved information. However, there are cases where answering a question requires implicit knowledge that is not directly retrievable from the question itself. In this work, we propose a novel question-answering pipeline called BeamSearchQA. Our approach leverages large language models to iteratively generate new questions about the original question, enabling an iterative reasoning process. By iteratively refining and expanding the scope of the question, our method aims to capture and utilize hidden knowledge that may not be directly obtainable through retrieval. We evaluate our approach on the widely-used open-domain NQ and WebQ datasets. The experimental results demonstrate that BeamSearchQA significantly outperforms other zero-shot baselines, indicating its effectiveness in tackling the challenges of open-domain question answering.
comment: Work in progress
♻ ☆ How to Unleash the Power of Large Language Models for Few-shot Relation Extraction? ACL 2023
Scaling language models have revolutionized widespread NLP tasks, yet little comprehensively explored few-shot relation extraction with large language models. In this paper, we investigate principal methodologies, in-context learning and data generation, for few-shot relation extraction via GPT-3.5 through exhaustive experiments. To enhance few-shot performance, we further propose task-related instructions and schema-constrained data generation. We observe that in-context learning can achieve performance on par with previous prompt learning approaches, and data generation with the large language model can boost previous solutions to obtain new state-of-the-art few-shot results on four widely-studied relation extraction datasets. We hope our work can inspire future research for the capabilities of large language models in few-shot relation extraction. Code is available in https://github.com/zjunlp/DeepKE/tree/main/example/llm.
comment: SustaiNLP Workshop@ACL 2023
♻ ☆ Aerial Vision-and-Dialog Navigation ACL 2023
The ability to converse with humans and follow natural language commands is crucial for intelligent unmanned aerial vehicles (a.k.a. drones). It can relieve people's burden of holding a controller all the time, allow multitasking, and make drone control more accessible for people with disabilities or with their hands occupied. To this end, we introduce Aerial Vision-and-Dialog Navigation (AVDN), to navigate a drone via natural language conversation. We build a drone simulator with a continuous photorealistic environment and collect a new AVDN dataset of over 3k recorded navigation trajectories with asynchronous human-human dialogs between commanders and followers. The commander provides initial navigation instruction and further guidance by request, while the follower navigates the drone in the simulator and asks questions when needed. During data collection, followers' attention on the drone's visual observation is also recorded. Based on the AVDN dataset, we study the tasks of aerial navigation from (full) dialog history and propose an effective Human Attention Aided Transformer model (HAA-Transformer), which learns to predict both navigation waypoints and human attention.
comment: Accepted by ACL 2023 Findings
♻ ☆ A Global Context Mechanism for Sequence Labeling
Sequential labeling tasks necessitate the computation of sentence representations for each word within a given sentence. With the advent of advanced pretrained language models; one common approach involves incorporating a BiLSTM layer to bolster the sequence structure information at the output level. Nevertheless, it has been empirically demonstrated (P.-H. Li et al., 2020) that the potential of BiLSTM for generating sentence representations for sequence labeling tasks is constrained, primarily due to the amalgamation of fragments form past and future sentence representations to form a complete sentence representation. In this study, we discovered that strategically integrating the whole sentence representation, which existing in the first cell and last cell of BiLSTM, into sentence representation of ecah cell, could markedly enhance the F1 score and accuracy. Using BERT embedded within BiLSTM as illustration, we conducted exhaustive experiments on nine datasets for sequence labeling tasks, encompassing named entity recognition (NER), part of speech (POS) tagging and End-to-End Aspect-Based sentiment analysis (E2E-ABSA). We noted significant improvements in F1 scores and accuracy across all examined datasets .
♻ ☆ Reward Gaming in Conditional Text Generation ACL 2023
To align conditional text generation model outputs with desired behaviors, there has been an increasing focus on training the model using reinforcement learning (RL) with reward functions learned from human annotations. Under this framework, we identify three common cases where high rewards are incorrectly assigned to undesirable patterns: noise-induced spurious correlation, naturally occurring spurious correlation, and covariate shift. We show that even though learned metrics achieve high performance on the distribution of the data used to train the reward function, the undesirable patterns may be amplified during RL training of the text generation model. While there has been discussion about reward gaming in the RL or safety community, in this discussion piece, we would like to highlight reward gaming in the natural language generation (NLG) community using concrete conditional text generation examples and discuss potential fixes and areas for future work.
comment: ACL 2023
♻ ☆ Automatic Creation of Named Entity Recognition Datasets by Querying Phrase Representations ACL 2023
Most weakly supervised named entity recognition (NER) models rely on domain-specific dictionaries provided by experts. This approach is infeasible in many domains where dictionaries do not exist. While a phrase retrieval model was used to construct pseudo-dictionaries with entities retrieved from Wikipedia automatically in a recent study, these dictionaries often have limited coverage because the retriever is likely to retrieve popular entities rather than rare ones. In this study, we present a novel framework, HighGEN, that generates NER datasets with high-coverage pseudo-dictionaries. Specifically, we create entity-rich dictionaries with a novel search method, called phrase embedding search, which encourages the retriever to search a space densely populated with various entities. In addition, we use a new verification process based on the embedding distance between candidate entity mentions and entity types to reduce the false-positive noise in weak labels generated by high-coverage dictionaries. We demonstrate that HighGEN outperforms the previous best model by an average F1 score of 4.7 across five NER benchmark datasets.
comment: ACL 2023
♻ ☆ GENEVA: Benchmarking Generalizability for Event Argument Extraction with Hundreds of Event Types and Argument Roles ACL 2023
Recent works in Event Argument Extraction (EAE) have focused on improving model generalizability to cater to new events and domains. However, standard benchmarking datasets like ACE and ERE cover less than 40 event types and 25 entity-centric argument roles. Limited diversity and coverage hinder these datasets from adequately evaluating the generalizability of EAE models. In this paper, we first contribute by creating a large and diverse EAE ontology. This ontology is created by transforming FrameNet, a comprehensive semantic role labeling (SRL) dataset for EAE, by exploiting the similarity between these two tasks. Then, exhaustive human expert annotations are collected to build the ontology, concluding with 115 events and 220 argument roles, with a significant portion of roles not being entities. We utilize this ontology to further introduce GENEVA, a diverse generalizability benchmarking dataset comprising four test suites, aimed at evaluating models' ability to handle limited data and unseen event type generalization. We benchmark six EAE models from various families. The results show that owing to non-entity argument roles, even the best-performing model can only achieve 39% F1 score, indicating how GENEVA provides new challenges for generalization in EAE. Overall, our large and diverse EAE ontology can aid in creating more comprehensive future resources, while GENEVA is a challenging benchmarking dataset encouraging further research for improving generalizability in EAE. The code and data can be found at https://github.com/PlusLabNLP/GENEVA.
comment: Accepted at ACL 2023 main conference
♻ ☆ Level Generation Through Large Language Models
Large Language Models (LLMs) are powerful tools, capable of leveraging their training on natural language to write stories, generate code, and answer questions. But can they generate functional video game levels? Game levels, with their complex functional constraints and spatial relationships in more than one dimension, are very different from the kinds of data an LLM typically sees during training. Datasets of game levels are also hard to come by, potentially taxing the abilities of these data-hungry models. We investigate the use of LLMs to generate levels for the game Sokoban, finding that LLMs are indeed capable of doing so, and that their performance scales dramatically with dataset size. We also perform preliminary experiments on controlling LLM level generators and discuss promising areas for future work.
♻ ☆ Towards Understanding Chain-of-Thought Prompting: An Empirical Study of What Matters ACL-23
Chain-of-Thought (CoT) prompting can dramatically improve the multi-step reasoning abilities of large language models (LLMs). CoT explicitly encourages the LLM to generate intermediate rationales for solving a problem, by providing a series of reasoning steps in the demonstrations. Despite its success, there is still little understanding of what makes CoT prompting effective and which aspects of the demonstrated reasoning steps contribute to its performance. In this paper, we show that CoT reasoning is possible even with invalid demonstrations - prompting with invalid reasoning steps can achieve over 80-90% of the performance obtained using CoT under various metrics, while still generating coherent lines of reasoning during inference. Further experiments show that other aspects of the rationales, such as being relevant to the query and correctly ordering the reasoning steps, are much more important for effective CoT reasoning. Overall, these findings both deepen our understanding of CoT prompting, and open up new questions regarding LLMs' capability to learn to reason in context.
comment: ACL-23 Camera Ready. Code and model input/output are available at https://github.com/sunlab-osu/Understanding-CoT
♻ ☆ On the Off-Target Problem of Zero-Shot Multilingual Neural Machine Translation ACL 2023
While multilingual neural machine translation has achieved great success, it suffers from the off-target issue, where the translation is in the wrong language. This problem is more pronounced on zero-shot translation tasks. In this work, we find that failing in encoding discriminative target language signal will lead to off-target and a closer lexical distance (i.e., KL-divergence) between two languages' vocabularies is related with a higher off-target rate. We also find that solely isolating the vocab of different languages in the decoder can alleviate the problem. Motivated by the findings, we propose Language Aware Vocabulary Sharing (LAVS), a simple and effective algorithm to construct the multilingual vocabulary, that greatly alleviates the off-target problem of the translation model by increasing the KL-divergence between languages. We conduct experiments on a multilingual machine translation benchmark in 11 languages. Experiments show that the off-target rate for 90 translation tasks is reduced from 29\% to 8\%, while the overall BLEU score is improved by an average of 1.9 points without extra training cost or sacrificing the supervised directions' performance. We release the code at \href{https://github.com/chenllliang/Off-Target-MNMT}{https://github.com/chenllliang/Off-Target-MNMT} for reproduction.
comment: Findings of ACL 2023
♻ ☆ FACTUAL: A Benchmark for Faithful and Consistent Textual Scene Graph Parsing ACL 2023
Textual scene graph parsing has become increasingly important in various vision-language applications, including image caption evaluation and image retrieval. However, existing scene graph parsers that convert image captions into scene graphs often suffer from two types of errors. First, the generated scene graphs fail to capture the true semantics of the captions or the corresponding images, resulting in a lack of faithfulness. Second, the generated scene graphs have high inconsistency, with the same semantics represented by different annotations. To address these challenges, we propose a novel dataset, which involves re-annotating the captions in Visual Genome (VG) using a new intermediate representation called FACTUAL-MR. FACTUAL-MR can be directly converted into faithful and consistent scene graph annotations. Our experimental results clearly demonstrate that the parser trained on our dataset outperforms existing approaches in terms of faithfulness and consistency. This improvement leads to a significant performance boost in both image caption evaluation and zero-shot image retrieval tasks. Furthermore, we introduce a novel metric for measuring scene graph similarity, which, when combined with the improved scene graph parser, achieves state-of-the-art (SOTA) results on multiple benchmark datasets for the aforementioned tasks. The code and dataset are available at https://github.com/zhuang-li/FACTUAL .
comment: 9 pages, ACL 2023 (findings)
♻ ☆ Large Language Models are Frame-level Directors for Zero-shot Text-to-Video Generation
In the paradigm of AI-generated content (AIGC), there has been increasing attention in extending pre-trained text-to-image (T2I) models to text-to-video (T2V) generation. Despite their effectiveness, these frameworks face challenges in maintaining consistent narratives and handling rapid shifts in scene composition or object placement from a single user prompt. This paper introduces a new framework, dubbed DirecT2V, which leverages instruction-tuned large language models (LLMs) to generate frame-by-frame descriptions from a single abstract user prompt. DirecT2V utilizes LLM directors to divide user inputs into separate prompts for each frame, enabling the inclusion of time-varying content and facilitating consistent video generation. To maintain temporal consistency and prevent object collapse, we propose a novel value mapping method and dual-softmax filtering. Extensive experimental results validate the effectiveness of the DirecT2V framework in producing visually coherent and consistent videos from abstract user prompts, addressing the challenges of zero-shot video generation.
comment: The code and demo will be available at https://github.com/KU-CVLAB/DirecT2V
♻ ☆ Super-CLEVR: A Virtual Benchmark to Diagnose Domain Robustness in Visual Reasoning CVPR 2023
Visual Question Answering (VQA) models often perform poorly on out-of-distribution data and struggle on domain generalization. Due to the multi-modal nature of this task, multiple factors of variation are intertwined, making generalization difficult to analyze. This motivates us to introduce a virtual benchmark, Super-CLEVR, where different factors in VQA domain shifts can be isolated in order that their effects can be studied independently. Four factors are considered: visual complexity, question redundancy, concept distribution and concept compositionality. With controllably generated data, Super-CLEVR enables us to test VQA methods in situations where the test data differs from the training data along each of these axes. We study four existing methods, including two neural symbolic methods NSCL and NSVQA, and two non-symbolic methods FiLM and mDETR; and our proposed method, probabilistic NSVQA (P-NSVQA), which extends NSVQA with uncertainty reasoning. P-NSVQA outperforms other methods on three of the four domain shift factors. Our results suggest that disentangling reasoning and perception, combined with probabilistic uncertainty, form a strong VQA model that is more robust to domain shifts. The dataset and code are released at https://github.com/Lizw14/Super-CLEVR.
comment: Published in CVPR 2023 as Highlight. Data and code are released at https://github.com/Lizw14/Super-CLEVR
♻ ☆ Communication Drives the Emergence of Language Universals in Neural Agents: Evidence from the Word-order/Case-marking Trade-off ACL
Artificial learners often behave differently from human learners in the context of neural agent-based simulations of language emergence and change. A common explanation is the lack of appropriate cognitive biases in these learners. However, it has also been proposed that more naturalistic settings of language learning and use could lead to more human-like results. We investigate this latter account focusing on the word-order/case-marking trade-off, a widely attested language universal that has proven particularly hard to simulate. We propose a new Neural-agent Language Learning and Communication framework (NeLLCom) where pairs of speaking and listening agents first learn a miniature language via supervised learning, and then optimize it for communication via reinforcement learning. Following closely the setup of earlier human experiments, we succeed in replicating the trade-off with the new framework without hard-coding specific biases in the agents. We see this as an essential step towards the investigation of language universals with neural learners.
comment: Accepted to TACL, pre-MIT Press publication version
♻ ☆ Query-Utterance Attention with Joint modeling for Query-Focused Meeting Summarization ICASSP 2023
Query-focused meeting summarization (QFMS) aims to generate summaries from meeting transcripts in response to a given query. Previous works typically concatenate the query with meeting transcripts and implicitly model the query relevance only at the token level with attention mechanism. However, due to the dilution of key query-relevant information caused by long meeting transcripts, the original transformer-based model is insufficient to highlight the key parts related to the query. In this paper, we propose a query-aware framework with joint modeling token and utterance based on Query-Utterance Attention. It calculates the utterance-level relevance to the query with a dense retrieval module. Then both token-level query relevance and utterance-level query relevance are combined and incorporated into the generation process with attention mechanism explicitly. We show that the query relevance of different granularities contributes to generating a summary more related to the query. Experimental results on the QMSum dataset show that the proposed model achieves new state-of-the-art performance.
comment: Accepted by ICASSP 2023
♻ ☆ A Pilot Study on Dialogue-Level Dependency Parsing for Chinese ACL 2023
Dialogue-level dependency parsing has received insufficient attention, especially for Chinese. To this end, we draw on ideas from syntactic dependency and rhetorical structure theory (RST), developing a high-quality human-annotated corpus, which contains 850 dialogues and 199,803 dependencies. Considering that such tasks suffer from high annotation costs, we investigate zero-shot and few-shot scenarios. Based on an existing syntactic treebank, we adopt a signal-based method to transform seen syntactic dependencies into unseen ones between elementary discourse units (EDUs), where the signals are detected by masked language modeling. Besides, we apply single-view and multi-view data selection to access reliable pseudo-labeled instances. Experimental results show the effectiveness of these baselines. Moreover, we discuss several crucial points about our dataset and approach.
comment: Accepted by Findings of ACL 2023 (Camera-ready version)
♻ ☆ Do Large Language Models know what humans know?
Humans can attribute beliefs to others. However, it is unknown to what extent this ability results from an innate biological endowment or from experience accrued through child development, particularly exposure to language describing others' mental states. We test the viability of the language exposure hypothesis by assessing whether models exposed to large quantities of human language display sensitivity to the implied knowledge states of characters in written passages. In pre-registered analyses, we present a linguistic version of the False Belief Task to both human participants and a Large Language Model, GPT-3. Both are sensitive to others' beliefs, but while the language model significantly exceeds chance behavior, it does not perform as well as the humans, nor does it explain the full extent of their behavior -- despite being exposed to more language than a human would in a lifetime. This suggests that while statistical learning from language exposure may in part explain how humans develop the ability to reason about the mental states of others, other mechanisms are also responsible.
♻ ☆ Bridging the Gap: A Survey on Integrating (Human) Feedback for Natural Language Generation
Many recent advances in natural language generation have been fueled by training large language models on internet-scale data. However, this paradigm can lead to models that generate toxic, inaccurate, and unhelpful content, and automatic evaluation metrics often fail to identify these behaviors. As models become more capable, human feedback is an invaluable signal for evaluating and improving models. This survey aims to provide an overview of the recent research that has leveraged human feedback to improve natural language generation. First, we introduce an encompassing formalization of feedback, and identify and organize existing research into a taxonomy following this formalization. Next, we discuss how feedback can be described by its format and objective, and cover the two approaches proposed to use feedback (either for training or decoding): directly using the feedback or training feedback models. We also discuss existing datasets for human-feedback data collection, and concerns surrounding feedback collection. Finally, we provide an overview of the nascent field of AI feedback, which exploits large language models to make judgments based on a set of principles and minimize the need for human intervention.
comment: Work in Progress
♻ ☆ N-best T5: Robust ASR Error Correction using Multiple Input Hypotheses and Constrained Decoding Space INTERSPEECH
Error correction models form an important part of Automatic Speech Recognition (ASR) post-processing to improve the readability and quality of transcriptions. Most prior works use the 1-best ASR hypothesis as input and therefore can only perform correction by leveraging the context within one sentence. In this work, we propose a novel N-best T5 model for this task, which is fine-tuned from a T5 model and utilizes ASR N-best lists as model input. By transferring knowledge from the pre-trained language model and obtaining richer information from the ASR decoding space, the proposed approach outperforms a strong Conformer-Transducer baseline. Another issue with standard error correction is that the generation process is not well-guided. To address this a constrained decoding process, either based on the N-best list or an ASR lattice, is used which allows additional information to be propagated.
comment: submitted to INTERSPEECH
♻ ☆ Quick Dense Retrievers Consume KALE: Post Training Kullback Leibler Alignment of Embeddings for Asymmetrical dual encoders ACL 2023
In this paper, we consider the problem of improving the inference latency of language model-based dense retrieval systems by introducing structural compression and model size asymmetry between the context and query encoders. First, we investigate the impact of pre and post-training compression on the MSMARCO, Natural Questions, TriviaQA, SQUAD, and SCIFACT, finding that asymmetry in the dual encoders in dense retrieval can lead to improved inference efficiency. Knowing this, we introduce Kullback Leibler Alignment of Embeddings (KALE), an efficient and accurate method for increasing the inference efficiency of dense retrieval methods by pruning and aligning the query encoder after training. Specifically, KALE extends traditional Knowledge Distillation after bi-encoder training, allowing for effective query encoder compression without full retraining or index generation. Using KALE and asymmetric training, we can generate models which exceed the performance of DistilBERT despite having 3x faster inference.
comment: SustaiNLP2023 @ ACL 2023, 8 pages, 4 figures, 30 tables
♻ ☆ SenteCon: Leveraging Lexicons to Learn Human-Interpretable Language Representations ACL 2023
Although deep language representations have become the dominant form of language featurization in recent years, in many settings it is important to understand a model's decision-making process. This necessitates not only an interpretable model but also interpretable features. In particular, language must be featurized in a way that is interpretable while still characterizing the original text well. We present SenteCon, a method for introducing human interpretability in deep language representations. Given a passage of text, SenteCon encodes the text as a layer of interpretable categories in which each dimension corresponds to the relevance of a specific category. Our empirical evaluations indicate that encoding language with SenteCon provides high-level interpretability at little to no cost to predictive performance on downstream tasks. Moreover, we find that SenteCon outperforms existing interpretable language representations with respect to both its downstream performance and its agreement with human characterizations of the text.
comment: Accepted to Findings of ACL 2023
♻ ☆ Faith and Fate: Limits of Transformers on Compositionality
Transformer large language models (LLMs) have sparked admiration for their exceptional performance on tasks that demand intricate multi-step reasoning. Yet, these models simultaneously show failures on surprisingly trivial problems. This begs the question: Are these errors incidental, or do they signal more substantial limitations? In an attempt to demystify Transformers, we investigate the limits of these models across three representative compositional tasks -- multi-digit multiplication, logic grid puzzles, and a classic dynamic programming problem. These tasks require breaking problems down into sub-steps and synthesizing these steps into a precise answer. We formulate compositional tasks as computation graphs to systematically quantify the level of complexity, and break down reasoning steps into intermediate sub-procedures. Our empirical findings suggest that Transformers solve compositional tasks by reducing multi-step compositional reasoning into linearized subgraph matching, without necessarily developing systematic problem-solving skills. To round off our empirical study, we provide theoretical arguments on abstract multi-step reasoning problems that highlight how Transformers' performance will rapidly decay with increased task complexity.
comment: 10 pages + appendix (21 pages)
♻ ☆ "I'm fully who I am": Towards Centering Transgender and Non-Binary Voices to Measure Biases in Open Language Generation
Transgender and non-binary (TGNB) individuals disproportionately experience discrimination and exclusion from daily life. Given the recent popularity and adoption of language generation technologies, the potential to further marginalize this population only grows. Although a multitude of NLP fairness literature focuses on illuminating and addressing gender biases, assessing gender harms for TGNB identities requires understanding how such identities uniquely interact with societal gender norms and how they differ from gender binary-centric perspectives. Such measurement frameworks inherently require centering TGNB voices to help guide the alignment between gender-inclusive NLP and whom they are intended to serve. Towards this goal, we ground our work in the TGNB community and existing interdisciplinary literature to assess how the social reality surrounding experienced marginalization of TGNB persons contributes to and persists within Open Language Generation (OLG). This social knowledge serves as a guide for evaluating popular large language models (LLMs) on two key aspects: (1) misgendering and (2) harmful responses to gender disclosure. To do this, we introduce TANGO, a dataset of template-based real-world text curated from a TGNB-oriented community. We discover a dominance of binary gender norms reflected by the models; LLMs least misgendered subjects in generated text when triggered by prompts whose subjects used binary pronouns. Meanwhile, misgendering was most prevalent when triggering generation with singular they and neopronouns. When prompted with gender disclosures, TGNB disclosure generated the most stigmatizing language and scored most toxic, on average. Our findings warrant further research on how TGNB harms manifest in LLMs and serve as a broader case study toward concretely grounding the design of gender-inclusive AI in community voices and interdisciplinary literature.
♻ ☆ On the Possibilities of AI-Generated Text Detection
Our work focuses on the challenge of detecting outputs generated by Large Language Models (LLMs) to distinguish them from those generated by humans. This ability is of the utmost importance in numerous applications. However, the possibility of such discernment has been the subject of debate within the community. Therefore, a central question is whether we can detect AI-generated text and, if so, when. In this work, we provide evidence that it should almost always be possible to detect AI-generated text unless the distributions of human and machine-generated texts are exactly the same over the entire support. This observation follows from the standard results in information theory and relies on the fact that if the machine text becomes more human-like, we need more samples to detect it. We derive a precise sample complexity bound of AI-generated text detection, which tells how many samples are needed to detect AI-generated text. This gives rise to additional challenges of designing more complicated detectors that take in $n$ samples for detection (rather than just one), which is the scope of future research on this topic. Our empirical evaluations on various real and synthetic datasets support our claim about the existence of better detectors, demonstrating that AI-generated text detection should be achievable in the majority of scenarios. Our theory and results align with OpenAI's empirical findings, (in relation to sequence length), and we are the first to provide a solid theoretical justification for these outcomes.
♻ ☆ Why Did the Chicken Cross the Road? Rephrasing and Analyzing Ambiguous Questions in VQA ACL 2023
Natural language is ambiguous. Resolving ambiguous questions is key to successfully answering them. Focusing on questions about images, we create a dataset of ambiguous examples. We annotate these, grouping answers by the underlying question they address and rephrasing the question for each group to reduce ambiguity. Our analysis reveals a linguistically-aligned ontology of reasons for ambiguity in visual questions. We then develop an English question-generation model which we demonstrate via automatic and human evaluation produces less ambiguous questions. We further show that the question generation objective we use allows the model to integrate answer group information without any direct supervision.
comment: ACL 2023. Code and data: https://github.com/esteng/ambiguous_vqa
♻ ☆ True Detective: A Deep Abductive Reasoning Benchmark Undoable for GPT-3 and Challenging for GPT-4
Large language models (LLMs) have demonstrated solid zero-shot reasoning capabilities, which is reflected in their performance on the current test tasks. This calls for a more challenging benchmark requiring highly advanced reasoning ability to be solved. In this paper, we introduce such a benchmark, consisting of 191 long-form (1200 words on average) mystery narratives constructed as detective puzzles. Puzzles are sourced from the "5 Minute Mystery" platform and include a multiple-choice question for evaluation. Only 47% of humans solve a puzzle successfully on average, while the best human solvers achieve over 80% success rate. We show that GPT-3 models barely outperform random on this benchmark (with 28% accuracy) while state-of-the-art GPT-4 solves only 38% of puzzles. This indicates that there is still a significant gap in the deep reasoning abilities of LLMs and humans and highlights the need for further research in this area. Our work introduces a challenging benchmark for future studies on reasoning in language models and contributes to a better understanding of the limits of LLMs' abilities.
comment: 5 pages, to appear at *SEM
♻ ☆ ProKnow: Process Knowledge for Safety Constrained and Explainable Question Generation for Mental Health Diagnostic Assistance
Current Virtual Mental Health Assistants (VMHAs) provide counseling and suggestive care. They refrain from patient diagnostic assistance because they lack training in safety-constrained and specialized clinical process knowledge. In this work, we define Proknow as an ordered set of information that maps to evidence-based guidelines or categories of conceptual understanding to experts in a domain. We also introduce a new dataset of diagnostic conversations guided by safety constraints and Proknow that healthcare professionals use. We develop a method for natural language question generation (NLG) that collects diagnostic information from the patient interactively. We demonstrate the limitations of using state-of-the-art large-scale language models (LMs) on this dataset. Our algorithm models the process knowledge through explicitly modeling safety, knowledge capture, and explainability. LMs augmented with ProKnow guided method generated 89% safer questions in the depression and anxiety domain. The Explainability of the generated question is assessed by computing similarity with concepts in depression and anxiety knowledge bases. Overall, irrespective of the type of LMs augmented with our ProKnow, we achieved an average 82% improvement over simple pre-trained LMs on safety, explainability, and process-guided question generation. We qualitatively and quantitatively evaluate the efficacy of the proposed ProKnow-guided methods by introducing three new evaluation metrics for safety, explainability, and process knowledge adherence.
Computer Vision and Pattern Recognition 167
☆ Hiera: A Hierarchical Vision Transformer without the Bells-and-Whistles ICML 2023
Modern hierarchical vision transformers have added several vision-specific components in the pursuit of supervised classification performance. While these components lead to effective accuracies and attractive FLOP counts, the added complexity actually makes these transformers slower than their vanilla ViT counterparts. In this paper, we argue that this additional bulk is unnecessary. By pretraining with a strong visual pretext task (MAE), we can strip out all the bells-and-whistles from a state-of-the-art multi-stage vision transformer without losing accuracy. In the process, we create Hiera, an extremely simple hierarchical vision transformer that is more accurate than previous models while being significantly faster both at inference and during training. We evaluate Hiera on a variety of tasks for image and video recognition. Our code and models are available at https://github.com/facebookresearch/hiera.
comment: ICML 2023 Oral version. Code+Models: https://github.com/facebookresearch/hiera
☆ StyleGAN knows Normal, Depth, Albedo, and More
Intrinsic images, in the original sense, are image-like maps of scene properties like depth, normal, albedo or shading. This paper demonstrates that StyleGAN can easily be induced to produce intrinsic images. The procedure is straightforward. We show that, if StyleGAN produces $G({w})$ from latents ${w}$, then for each type of intrinsic image, there is a fixed offset ${d}_c$ so that $G({w}+{d}_c)$ is that type of intrinsic image for $G({w})$. Here ${d}_c$ is {\em independent of ${w}$}. The StyleGAN we used was pretrained by others, so this property is not some accident of our training regime. We show that there are image transformations StyleGAN will {\em not} produce in this fashion, so StyleGAN is not a generic image regression engine. It is conceptually exciting that an image generator should ``know'' and represent intrinsic images. There may also be practical advantages to using a generative model to produce intrinsic images. The intrinsic images obtained from StyleGAN compare well both qualitatively and quantitatively with those obtained by using SOTA image regression techniques; but StyleGAN's intrinsic images are robust to relighting effects, unlike SOTA methods.
comment: Beyond Image Generation: StyleGAN knows Normals, Depth, Albedo, Shading, Segmentation and perhaps more!
☆ Continual Learning for Abdominal Multi-Organ and Tumor Segmentation MICCAI-2023
The ability to dynamically extend a model to new data and classes is critical for multiple organ and tumor segmentation. However, due to privacy regulations, accessing previous data and annotations can be problematic in the medical domain. This poses a significant barrier to preserving the high segmentation accuracy of the old classes when learning from new classes because of the catastrophic forgetting problem. In this paper, we first empirically demonstrate that simply using high-quality pseudo labels can fairly mitigate this problem in the setting of organ segmentation. Furthermore, we put forward an innovative architecture designed specifically for continuous organ and tumor segmentation, which incurs minimal computational overhead. Our proposed design involves replacing the conventional output layer with a suite of lightweight, class-specific heads, thereby offering the flexibility to accommodate newly emerging classes. These heads enable independent predictions for newly introduced and previously learned classes, effectively minimizing the impact of new classes on old ones during the course of continual learning. We further propose incorporating Contrastive Language-Image Pretraining (CLIP) embeddings into the organ-specific heads. These embeddings encapsulate the semantic information of each class, informed by extensive image-text co-training. The proposed method is evaluated on both in-house and public abdominal CT datasets under organ and tumor segmentation tasks. Empirical results suggest that the proposed design improves the segmentation performance of a baseline neural network on newly-introduced and previously-learned classes along the learning trajectory.
comment: MICCAI-2023
☆ Diffusion Self-Guidance for Controllable Image Generation
Large-scale generative models are capable of producing high-quality images from detailed text descriptions. However, many aspects of an image are difficult or impossible to convey through text. We introduce self-guidance, a method that provides greater control over generated images by guiding the internal representations of diffusion models. We demonstrate that properties such as the shape, location, and appearance of objects can be extracted from these representations and used to steer sampling. Self-guidance works similarly to classifier guidance, but uses signals present in the pretrained model itself, requiring no additional models or training. We show how a simple set of properties can be composed to perform challenging image manipulations, such as modifying the position or size of objects, merging the appearance of objects in one image with the layout of another, composing objects from many images into one, and more. We also show that self-guidance can be used to edit real images. For results and an interactive demo, see our project page at https://dave.ml/selfguidance/
comment: Project page at https://dave.ml/selfguidance/
☆ Using generative AI to investigate medical imagery models and datasets
AI models have shown promise in many medical imaging tasks. However, our ability to explain what signals these models have learned is severely lacking. Explanations are needed in order to increase the trust in AI-based models, and could enable novel scientific discovery by uncovering signals in the data that are not yet known to experts. In this paper, we present a method for automatic visual explanations leveraging team-based expertise by generating hypotheses of what visual signals in the images are correlated with the task. We propose the following 4 steps: (i) Train a classifier to perform a given task (ii) Train a classifier guided StyleGAN-based image generator (StylEx) (iii) Automatically detect and visualize the top visual attributes that the classifier is sensitive towards (iv) Formulate hypotheses for the underlying mechanisms, to stimulate future research. Specifically, we present the discovered attributes to an interdisciplinary panel of experts so that hypotheses can account for social and structural determinants of health. We demonstrate results on eight prediction tasks across three medical imaging modalities: retinal fundus photographs, external eye photographs, and chest radiographs. We showcase examples of attributes that capture clinically known features, confounders that arise from factors beyond physiological mechanisms, and reveal a number of physiologically plausible novel attributes. Our approach has the potential to enable researchers to better understand, improve their assessment, and extract new knowledge from AI-based models. Importantly, we highlight that attributes generated by our framework can capture phenomena beyond physiology or pathophysiology, reflecting the real world nature of healthcare delivery and socio-cultural factors. Finally, we intend to release code to enable researchers to train their own StylEx models and analyze their predictive tasks.
comment: 34 pages, 1 figure
☆ StyleDrop: Text-to-Image Generation in Any Style
Pre-trained large text-to-image models synthesize impressive images with an appropriate use of text prompts. However, ambiguities inherent in natural language and out-of-distribution effects make it hard to synthesize image styles, that leverage a specific design pattern, texture or material. In this paper, we introduce StyleDrop, a method that enables the synthesis of images that faithfully follow a specific style using a text-to-image model. The proposed method is extremely versatile and captures nuances and details of a user-provided style, such as color schemes, shading, design patterns, and local and global effects. It efficiently learns a new style by fine-tuning very few trainable parameters (less than $1\%$ of total model parameters) and improving the quality via iterative training with either human or automated feedback. Better yet, StyleDrop is able to deliver impressive results even when the user supplies only a single image that specifies the desired style. An extensive study shows that, for the task of style tuning text-to-image models, StyleDrop implemented on Muse convincingly outperforms other methods, including DreamBooth and textual inversion on Imagen or Stable Diffusion. More results are available at our project website: https://styledrop.github.io
comment: Preprint. Project page at https://styledrop.github.io
☆ StableRep: Synthetic Images from Text-to-Image Models Make Strong Visual Representation Learners
We investigate the potential of learning visual representations using synthetic images generated by text-to-image models. This is a natural question in the light of the excellent performance of such models in generating high-quality images. We consider specifically the Stable Diffusion, one of the leading open source text-to-image models. We show that (1) when the generative model is configured with proper classifier-free guidance scale, training self-supervised methods on synthetic images can match or beat the real image counterpart; (2) by treating the multiple images generated from the same text prompt as positives for each other, we develop a multi-positive contrastive learning method, which we call StableRep. With solely synthetic images, the representations learned by StableRep surpass the performance of representations learned by SimCLR and CLIP using the same set of text prompts and corresponding real images, on large scale datasets. When we further add language supervision, StableRep trained with 20M synthetic images achieves better accuracy than CLIP trained with 50M real images.
☆ SnapFusion: Text-to-Image Diffusion Model on Mobile Devices within Two Seconds
Text-to-image diffusion models can create stunning images from natural language descriptions that rival the work of professional artists and photographers. However, these models are large, with complex network architectures and tens of denoising iterations, making them computationally expensive and slow to run. As a result, high-end GPUs and cloud-based inference are required to run diffusion models at scale. This is costly and has privacy implications, especially when user data is sent to a third party. To overcome these challenges, we present a generic approach that, for the first time, unlocks running text-to-image diffusion models on mobile devices in less than $2$ seconds. We achieve so by introducing efficient network architecture and improving step distillation. Specifically, we propose an efficient UNet by identifying the redundancy of the original model and reducing the computation of the image decoder via data distillation. Further, we enhance the step distillation by exploring training strategies and introducing regularization from classifier-free guidance. Our extensive experiments on MS-COCO show that our model with $8$ denoising steps achieves better FID and CLIP scores than Stable Diffusion v$1.5$ with $50$ steps. Our work democratizes content creation by bringing powerful text-to-image diffusion models to the hands of users.
comment: Our project webpage: https://snap-research.github.io/SnapFusion/
☆ Building Rearticulable Models for Arbitrary 3D Objects from 4D Point Clouds CVPR 2023
We build rearticulable models for arbitrary everyday man-made objects containing an arbitrary number of parts that are connected together in arbitrary ways via 1 degree-of-freedom joints. Given point cloud videos of such everyday objects, our method identifies the distinct object parts, what parts are connected to what other parts, and the properties of the joints connecting each part pair. We do this by jointly optimizing the part segmentation, transformation, and kinematics using a novel energy minimization framework. Our inferred animatable models, enables retargeting to novel poses with sparse point correspondences guidance. We test our method on a new articulating robot dataset, and the Sapiens dataset with common daily objects, as well as real-world scans. Experiments show that our method outperforms two leading prior works on various metrics.
comment: Accepted to CVPR 2023. Project page: https://stevenlsw.github.io/reart
☆ AGILE3D: Attention Guided Interactive Multi-object 3D Segmentation
During interactive segmentation, a model and a user work together to delineate objects of interest in a 3D point cloud. In an iterative process, the model assigns each data point to an object (or the background), while the user corrects errors in the resulting segmentation and feeds them back into the model. From a machine learning perspective the goal is to design the model and the feedback mechanism in a way that minimizes the required user input. The current best practice segments objects one at a time, and asks the user to provide positive clicks to indicate regions wrongly assigned to the background and negative clicks to indicate regions wrongly assigned to the object (foreground). Sequentially visiting objects is wasteful, since it disregards synergies between objects: a positive click for a given object can, by definition, serve as a negative click for nearby objects, moreover a direct competition between adjacent objects can speed up the identification of their common boundary. We introduce AGILE3D, an efficient, attention-based model that (1) supports simultaneous segmentation of multiple 3D objects, (2) yields more accurate segmentation masks with fewer user clicks, and (3) offers faster inference. We encode the point cloud into a latent feature representation, and view user clicks as queries and employ cross-attention to represent contextual relations between different click locations as well as between clicks and the 3D point cloud features. Every time new clicks are added, we only need to run a lightweight decoder that produces updated segmentation masks. In experiments with four different point cloud datasets, AGILE3D sets a new state of the art, moreover, we also verify its practicality in real-world setups with a real user study.
comment: Project page: https://ywyue.github.io/AGILE3D/
☆ Active Reinforcement Learning under Limited Visual Observability
In this work, we investigate Active Reinforcement Learning (Active-RL), where an embodied agent simultaneously learns action policy for the task while also controlling its visual observations in partially observable environments. We denote the former as motor policy and the latter as sensory policy. For example, humans solve real world tasks by hand manipulation (motor policy) together with eye movements (sensory policy). Active-RL poses challenges on coordinating two policies given their mutual influence. We propose SUGARL, Sensorimotor Understanding Guided Active Reinforcement Learning, a framework that models motor and sensory policies separately, but jointly learns them using with an intrinsic sensorimotor reward. This learnable reward is assigned by sensorimotor reward module, incentivizes the sensory policy to select observations that are optimal to infer its own motor action, inspired by the sensorimotor stage of humans. Through a series of experiments, we show the effectiveness of our method across a range of observability conditions and its adaptability to existed RL algorithms. The sensory policies learned through our method are observed to exhibit effective active vision strategies.
comment: Project page at https://elicassion.github.io/sugarl/sugarl.html Code at https://github.com/elicassion/sugarl Environment library at https://github.com/elicassion/active-gym
☆ Intriguing Properties of Text-guided Diffusion Models
Text-guided diffusion models (TDMs) are widely applied but can fail unexpectedly. Common failures include: (i) natural-looking text prompts generating images with the wrong content, or (ii) different random samples of the latent variables that generate vastly different, and even unrelated, outputs despite being conditioned on the same text prompt. In this work, we aim to study and understand the failure modes of TDMs in more detail. To achieve this, we propose SAGE, an adversarial attack on TDMs that uses image classifiers as surrogate loss functions, to search over the discrete prompt space and the high-dimensional latent space of TDMs to automatically discover unexpected behaviors and failure cases in the image generation. We make several technical contributions to ensure that SAGE finds failure cases of the diffusion model, rather than the classifier, and verify this in a human study. Our study reveals four intriguing properties of TDMs that have not been systematically studied before: (1) We find a variety of natural text prompts producing images that fail to capture the semantics of input texts. We categorize these failures into ten distinct types based on the underlying causes. (2) We find samples in the latent space (which are not outliers) that lead to distorted images independent of the text prompt, suggesting that parts of the latent space are not well-structured. (3) We also find latent samples that lead to natural-looking images which are unrelated to the text prompt, implying a potential misalignment between the latent and prompt spaces. (4) By appending a single adversarial token embedding to an input prompt we can generate a variety of specified target objects, while only minimally affecting the CLIP score. This demonstrates the fragility of language representations and raises potential safety concerns.
comment: Code will be available at: https://github.com/qihao067/SAGE/
☆ Intelligent Grimm -- Open-ended Visual Storytelling via Latent Diffusion Models
Generative models have recently exhibited exceptional capabilities in various scenarios, for example, image generation based on text description. In this work, we focus on the task of generating a series of coherent image sequence based on a given storyline, denoted as open-ended visual storytelling. We make the following three contributions: (i) to fulfill the task of visual storytelling, we introduce two modules into a pre-trained stable diffusion model, and construct an auto-regressive image generator, termed as StoryGen, that enables to generate the current frame by conditioning on both a text prompt and a preceding frame; (ii) to train our proposed model, we collect paired image and text samples by sourcing from various online sources, such as videos, E-books, and establish a data processing pipeline for constructing a diverse dataset, named StorySalon, with a far larger vocabulary than existing animation-specific datasets; (iii) we adopt a three-stage curriculum training strategy, that enables style transfer, visual context conditioning, and human feedback alignment, respectively. Quantitative experiments and human evaluation have validated the superiority of our proposed model, in terms of image quality, style consistency, content consistency, and visual-language alignment. We will make the code, model, and dataset publicly available to the research community.
comment: Technical report
☆ ViCo: Detail-Preserving Visual Condition for Personalized Text-to-Image Generation
Personalized text-to-image generation using diffusion models has recently been proposed and attracted lots of attention. Given a handful of images containing a novel concept (e.g., a unique toy), we aim to tune the generative model to capture fine visual details of the novel concept and generate photorealistic images following a text condition. We present a plug-in method, named ViCo, for fast and lightweight personalized generation. Specifically, we propose an image attention module to condition the diffusion process on the patch-wise visual semantics. We introduce an attention-based object mask that comes almost at no cost from the attention module. In addition, we design a simple regularization based on the intrinsic properties of text-image attention maps to alleviate the common overfitting degradation. Unlike many existing models, our method does not finetune any parameters of the original diffusion model. This allows more flexible and transferable model deployment. With only light parameter training (~6% of the diffusion U-Net), our method achieves comparable or even better performance than all state-of-the-art models both qualitatively and quantitatively.
comment: Under review
☆ GRES: Generalized Referring Expression Segmentation CVPR2023
Referring Expression Segmentation (RES) aims to generate a segmentation mask for the object described by a given language expression. Existing classic RES datasets and methods commonly support single-target expressions only, i.e., one expression refers to one target object. Multi-target and no-target expressions are not considered. This limits the usage of RES in practice. In this paper, we introduce a new benchmark called Generalized Referring Expression Segmentation (GRES), which extends the classic RES to allow expressions to refer to an arbitrary number of target objects. Towards this, we construct the first large-scale GRES dataset called gRefCOCO that contains multi-target, no-target, and single-target expressions. GRES and gRefCOCO are designed to be well-compatible with RES, facilitating extensive experiments to study the performance gap of the existing RES methods on the GRES task. In the experimental study, we find that one of the big challenges of GRES is complex relationship modeling. Based on this, we propose a region-based GRES baseline ReLA that adaptively divides the image into regions with sub-instance clues, and explicitly models the region-region and region-language dependencies. The proposed approach ReLA achieves new state-of-the-art performance on the both newly proposed GRES and classic RES tasks. The proposed gRefCOCO dataset and method are available at https://henghuiding.github.io/GRES.
comment: CVPR2023 Highlight, Project Page: https://henghuiding.github.io/GRES/
☆ The Hidden Language of Diffusion Models
Text-to-image diffusion models have demonstrated an unparalleled ability to generate high-quality, diverse images from a textual concept (e.g., "a doctor", "love"). However, the internal process of mapping text to a rich visual representation remains an enigma. In this work, we tackle the challenge of understanding concept representations in text-to-image models by decomposing an input text prompt into a small set of interpretable elements. This is achieved by learning a pseudo-token that is a sparse weighted combination of tokens from the model's vocabulary, with the objective of reconstructing the images generated for the given concept. Applied over the state-of-the-art Stable Diffusion model, this decomposition reveals non-trivial and surprising structures in the representations of concepts. For example, we find that some concepts such as "a president" or "a composer" are dominated by specific instances (e.g., "Obama", "Biden") and their interpolations. Other concepts, such as "happiness" combine associated terms that can be concrete ("family", "laughter") or abstract ("friendship", "emotion"). In addition to peering into the inner workings of Stable Diffusion, our method also enables applications such as single-image decomposition to tokens, bias detection and mitigation, and semantic image manipulation. Our code will be available at: https://hila-chefer.github.io/Conceptor/
☆ BUOL: A Bottom-Up Framework with Occupancy-aware Lifting for Panoptic 3D Scene Reconstruction From A Single Image CVPR 2023
Understanding and modeling the 3D scene from a single image is a practical problem. A recent advance proposes a panoptic 3D scene reconstruction task that performs both 3D reconstruction and 3D panoptic segmentation from a single image. Although having made substantial progress, recent works only focus on top-down approaches that fill 2D instances into 3D voxels according to estimated depth, which hinders their performance by two ambiguities. (1) instance-channel ambiguity: The variable ids of instances in each scene lead to ambiguity during filling voxel channels with 2D information, confusing the following 3D refinement. (2) voxel-reconstruction ambiguity: 2D-to-3D lifting with estimated single view depth only propagates 2D information onto the surface of 3D regions, leading to ambiguity during the reconstruction of regions behind the frontal view surface. In this paper, we propose BUOL, a Bottom-Up framework with Occupancy-aware Lifting to address the two issues for panoptic 3D scene reconstruction from a single image. For instance-channel ambiguity, a bottom-up framework lifts 2D information to 3D voxels based on deterministic semantic assignments rather than arbitrary instance id assignments. The 3D voxels are then refined and grouped into 3D instances according to the predicted 2D instance centers. For voxel-reconstruction ambiguity, the estimated multi-plane occupancy is leveraged together with depth to fill the whole regions of things and stuff. Our method shows a tremendous performance advantage over state-of-the-art methods on synthetic dataset 3D-Front and real-world dataset Matterport3D. Code and models are available in https://github.com/chtsy/buol.
comment: CVPR 2023, https://github.com/chtsy/buol
☆ Cocktail: Mixing Multi-Modality Controls for Text-Conditional Image Generation
Text-conditional diffusion models are able to generate high-fidelity images with diverse contents. However, linguistic representations frequently exhibit ambiguous descriptions of the envisioned objective imagery, requiring the incorporation of additional control signals to bolster the efficacy of text-guided diffusion models. In this work, we propose Cocktail, a pipeline to mix various modalities into one embedding, amalgamated with a generalized ControlNet (gControlNet), a controllable normalisation (ControlNorm), and a spatial guidance sampling method, to actualize multi-modal and spatially-refined control for text-conditional diffusion models. Specifically, we introduce a hyper-network gControlNet, dedicated to the alignment and infusion of the control signals from disparate modalities into the pre-trained diffusion model. gControlNet is capable of accepting flexible modality signals, encompassing the simultaneous reception of any combination of modality signals, or the supplementary fusion of multiple modality signals. The control signals are then fused and injected into the backbone model according to our proposed ControlNorm. Furthermore, our advanced spatial guidance sampling methodology proficiently incorporates the control signal into the designated region, thereby circumventing the manifestation of undesired objects within the generated image. We demonstrate the results of our method in controlling various modalities, proving high-quality synthesis and fidelity to multiple external signals.
comment: Project Page: https://mhh0318.github.io/cocktail/
☆ The ObjectFolder Benchmark: Multisensory Learning with Neural and Real Objects CVPR 2023
We introduce the ObjectFolder Benchmark, a benchmark suite of 10 tasks for multisensory object-centric learning, centered around object recognition, reconstruction, and manipulation with sight, sound, and touch. We also introduce the ObjectFolder Real dataset, including the multisensory measurements for 100 real-world household objects, building upon a newly designed pipeline for collecting the 3D meshes, videos, impact sounds, and tactile readings of real-world objects. We conduct systematic benchmarking on both the 1,000 multisensory neural objects from ObjectFolder, and the real multisensory data from ObjectFolder Real. Our results demonstrate the importance of multisensory perception and reveal the respective roles of vision, audio, and touch for different object-centric learning tasks. By publicly releasing our dataset and benchmark suite, we hope to catalyze and enable new research in multisensory object-centric learning in computer vision, robotics, and beyond. Project page: https://objectfolder.stanford.edu
comment: In CVPR 2023. Project page: https://objectfolder.stanford.edu/. ObjectFolder Real demo: https://www.objectfolder.org/swan_vis/. Gao, Dou, and Li contributed equally to this work
☆ Differential Diffusion: Giving Each Pixel Its Strength
Text-based image editing has advanced significantly in recent years. With the rise of diffusion models, image editing via textual instructions has become ubiquitous. Unfortunately, current models lack the ability to customize the quantity of the change per pixel or per image fragment, resorting to changing the entire image in an equal amount, or editing a specific region using a binary mask. In this paper, we suggest a new framework which enables the user to customize the quantity of change for each image fragment, thereby enhancing the flexibility and verbosity of modern diffusion models. Our framework does not require model training or fine-tuning, but instead performs everything at inference time, making it easily applicable to an existing model. We show both qualitatively and quantitatively that our method allows better controllability and can produce results which are unattainable by existing models. Our code is available at: https://github.com/exx8/differential-diffusion
comment: Our code is available at: https://github.com/exx8/differential-diffusion
☆ Make-Your-Video: Customized Video Generation Using Textual and Structural Guidance
Creating a vivid video from the event or scenario in our imagination is a truly fascinating experience. Recent advancements in text-to-video synthesis have unveiled the potential to achieve this with prompts only. While text is convenient in conveying the overall scene context, it may be insufficient to control precisely. In this paper, we explore customized video generation by utilizing text as context description and motion structure (e.g. frame-wise depth) as concrete guidance. Our method, dubbed Make-Your-Video, involves joint-conditional video generation using a Latent Diffusion Model that is pre-trained for still image synthesis and then promoted for video generation with the introduction of temporal modules. This two-stage learning scheme not only reduces the computing resources required, but also improves the performance by transferring the rich concepts available in image datasets solely into video generation. Moreover, we use a simple yet effective causal attention mask strategy to enable longer video synthesis, which mitigates the potential quality degradation effectively. Experimental results show the superiority of our method over existing baselines, particularly in terms of temporal coherence and fidelity to users' guidance. In addition, our model enables several intriguing applications that demonstrate potential for practical usage.
comment: 13 pages, 8 figures. Project page: https://doubiiu.github.io/projects/Make-Your-Video/
☆ Train Offline, Test Online: A Real Robot Learning Benchmark ICRA 2023
Three challenges limit the progress of robot learning research: robots are expensive (few labs can participate), everyone uses different robots (findings do not generalize across labs), and we lack internet-scale robotics data. We take on these challenges via a new benchmark: Train Offline, Test Online (TOTO). TOTO provides remote users with access to shared robotic hardware for evaluating methods on common tasks and an open-source dataset of these tasks for offline training. Its manipulation task suite requires challenging generalization to unseen objects, positions, and lighting. We present initial results on TOTO comparing five pretrained visual representations and four offline policy learning baselines, remotely contributed by five institutions. The real promise of TOTO, however, lies in the future: we release the benchmark for additional submissions from any user, enabling easy, direct comparison to several methods without the need to obtain hardware or collect data.
comment: Accepted to ICRA 2023
☆ "Let's not Quote out of Context": Unified Vision-Language Pretraining for Context Assisted Image Captioning
Well-formed context aware image captions and tags in enterprise content such as marketing material are critical to ensure their brand presence and content recall. Manual creation and updates to ensure the same is non trivial given the scale and the tedium towards this task. We propose a new unified Vision-Language (VL) model based on the One For All (OFA) model, with a focus on context-assisted image captioning where the caption is generated based on both the image and its context. Our approach aims to overcome the context-independent (image and text are treated independently) nature of the existing approaches. We exploit context by pretraining our model with datasets of three tasks: news image captioning where the news article is the context, contextual visual entailment, and keyword extraction from the context. The second pretraining task is a new VL task, and we construct and release two datasets for the task with 1.1M and 2.2K data instances. Our system achieves state-of-the-art results with an improvement of up to 8.34 CIDEr score on the benchmark news image captioning datasets. To the best of our knowledge, ours is the first effort at incorporating contextual information in pretraining the models for the VL tasks.
☆ Second Sight: Using brain-optimized encoding models to align image distributions with human brain activity
Two recent developments have accelerated progress in image reconstruction from human brain activity: large datasets that offer samples of brain activity in response to many thousands of natural scenes, and the open-sourcing of powerful stochastic image-generators that accept both low- and high-level guidance. Most work in this space has focused on obtaining point estimates of the target image, with the ultimate goal of approximating literal pixel-wise reconstructions of target images from the brain activity patterns they evoke. This emphasis belies the fact that there is always a family of images that are equally compatible with any evoked brain activity pattern, and the fact that many image-generators are inherently stochastic and do not by themselves offer a method for selecting the single best reconstruction from among the samples they generate. We introduce a novel reconstruction procedure (Second Sight) that iteratively refines an image distribution to explicitly maximize the alignment between the predictions of a voxel-wise encoding model and the brain activity patterns evoked by any target image. We show that our process converges on a distribution of high-quality reconstructions by refining both semantic content and low-level image details across iterations. Images sampled from these converged image distributions are competitive with state-of-the-art reconstruction algorithms. Interestingly, the time-to-convergence varies systematically across visual cortex, with earlier visual areas generally taking longer and converging on narrower image distributions, relative to higher-level brain areas. Second Sight thus offers a succinct and novel method for exploring the diversity of representations across visual brain areas.
comment: 15 Figures, 19 pages including the appendix
☆ Inserting Anybody in Diffusion Models via Celeb Basis
Exquisite demand exists for customizing the pretrained large text-to-image model, $\textit{e.g.}$, Stable Diffusion, to generate innovative concepts, such as the users themselves. However, the newly-added concept from previous customization methods often shows weaker combination abilities than the original ones even given several images during training. We thus propose a new personalization method that allows for the seamless integration of a unique individual into the pre-trained diffusion model using just $\textbf{one facial photograph}$ and only $\textbf{1024 learnable parameters}$ under $\textbf{3 minutes}$. So as we can effortlessly generate stunning images of this person in any pose or position, interacting with anyone and doing anything imaginable from text prompts. To achieve this, we first analyze and build a well-defined celeb basis from the embedding space of the pre-trained large text encoder. Then, given one facial photo as the target identity, we generate its own embedding by optimizing the weight of this basis and locking all other parameters. Empowered by the proposed celeb basis, the new identity in our customized model showcases a better concept combination ability than previous personalization methods. Besides, our model can also learn several new identities at once and interact with each other where the previous customization model fails to. The code will be released.
comment: Project page: http://celeb-basis.github.io ; Github repository: https://github.com/ygtxr1997/CelebBasis
☆ Sonicverse: A Multisensory Simulation Platform for Embodied Household Agents that See and Hear ICRA 2023
Developing embodied agents in simulation has been a key research topic in recent years. Exciting new tasks, algorithms, and benchmarks have been developed in various simulators. However, most of them assume deaf agents in silent environments, while we humans perceive the world with multiple senses. We introduce Sonicverse, a multisensory simulation platform with integrated audio-visual simulation for training household agents that can both see and hear. Sonicverse models realistic continuous audio rendering in 3D environments in real-time. Together with a new audio-visual VR interface that allows humans to interact with agents with audio, Sonicverse enables a series of embodied AI tasks that need audio-visual perception. For semantic audio-visual navigation in particular, we also propose a new multi-task learning model that achieves state-of-the-art performance. In addition, we demonstrate Sonicverse's realism via sim-to-real transfer, which has not been achieved by other simulators: an agent trained in Sonicverse can successfully perform audio-visual navigation in real-world environments. Sonicverse is available at: https://github.com/StanfordVL/Sonicverse.
comment: In ICRA 2023. Project page: https://ai.stanford.edu/~rhgao/sonicverse/. Code: https://github.com/StanfordVL/sonicverse. Gao and Li contributed equally to this work and are in alphabetical order
☆ Vocabulary-free Image Classification
Recent advances in large vision-language models have revolutionized the image classification paradigm. Despite showing impressive zero-shot capabilities, a pre-defined set of categories, a.k.a. the vocabulary, is assumed at test time for composing the textual prompts. However, such assumption can be impractical when the semantic context is unknown and evolving. We thus formalize a novel task, termed as Vocabulary-free Image Classification (VIC), where we aim to assign to an input image a class that resides in an unconstrained language-induced semantic space, without the prerequisite of a known vocabulary. VIC is a challenging task as the semantic space is extremely large, containing millions of concepts, with hard-to-discriminate fine-grained categories. In this work, we first empirically verify that representing this semantic space by means of an external vision-language database is the most effective way to obtain semantically relevant content for classifying the image. We then propose Category Search from External Databases (CaSED), a method that exploits a pre-trained vision-language model and an external vision-language database to address VIC in a training-free manner. CaSED first extracts a set of candidate categories from captions retrieved from the database based on their semantic similarity to the image, and then assigns to the image the best matching candidate category according to the same vision-language model. Experiments on benchmark datasets validate that CaSED outperforms other complex vision-language frameworks, while being efficient with much fewer parameters, paving the way for future research in this direction.
comment: 17 pages, 8 figures, code is available at https://github.com/altndrr/vic
☆ Conditioning Diffusion Models via Attributes and Semantic Masks for Face Generation
Deep generative models have shown impressive results in generating realistic images of faces. GANs managed to generate high-quality, high-fidelity images when conditioned on semantic masks, but they still lack the ability to diversify their output. Diffusion models partially solve this problem and are able to generate diverse samples given the same condition. In this paper, we propose a multi-conditioning approach for diffusion models via cross-attention exploiting both attributes and semantic masks to generate high-quality and controllable face images. We also studied the impact of applying perceptual-focused loss weighting into the latent space instead of the pixel space. Our method extends the previous approaches by introducing conditioning on more than one set of features, guaranteeing a more fine-grained control over the generated face images. We evaluate our approach on the CelebA-HQ dataset, and we show that it can generate realistic and diverse samples while allowing for fine-grained control over multiple attributes and semantic regions. Additionally, we perform an ablation study to evaluate the impact of different conditioning strategies on the quality and diversity of the generated images.
☆ MOSAIC: Masked Optimisation with Selective Attention for Image Reconstruction
Compressive sensing (CS) reconstructs images from sub-Nyquist measurements by solving a sparsity-regularized inverse problem. Traditional CS solvers use iterative optimizers with hand crafted sparsifiers, while early data-driven methods directly learn an inverse mapping from the low-dimensional measurement space to the original image space. The latter outperforms the former, but is restrictive to a pre-defined measurement domain. More recent, deep unrolling methods combine traditional proximal gradient methods and data-driven approaches to iteratively refine an image approximation. To achieve higher accuracy, it has also been suggested to learn both the sampling matrix, and the choice of measurement vectors adaptively. Contrary to the current trend, in this work we hypothesize that a general inverse mapping from a random set of compressed measurements to the image domain exists for a given measurement basis, and can be learned. Such a model is single-shot, non-restrictive and does not parametrize the sampling process. To this end, we propose MOSAIC, a novel compressive sensing framework to reconstruct images given any random selection of measurements, sampled using a fixed basis. Motivated by the uneven distribution of information across measurements, MOSAIC incorporates an embedding technique to efficiently apply attention mechanisms on an encoded sequence of measurements, while dispensing the need to use unrolled deep networks. A range of experiments validate our proposed architecture as a promising alternative for existing CS reconstruction methods, by achieving the state-of-the-art for metrics of reconstruction accuracy on standard datasets.
☆ T2IAT: Measuring Valence and Stereotypical Biases in Text-to-Image Generation ACL 2023
Warning: This paper contains several contents that may be toxic, harmful, or offensive. In the last few years, text-to-image generative models have gained remarkable success in generating images with unprecedented quality accompanied by a breakthrough of inference speed. Despite their rapid progress, human biases that manifest in the training examples, particularly with regard to common stereotypical biases, like gender and skin tone, still have been found in these generative models. In this work, we seek to measure more complex human biases exist in the task of text-to-image generations. Inspired by the well-known Implicit Association Test (IAT) from social psychology, we propose a novel Text-to-Image Association Test (T2IAT) framework that quantifies the implicit stereotypes between concepts and valence, and those in the images. We replicate the previously documented bias tests on generative models, including morally neutral tests on flowers and insects as well as demographic stereotypical tests on diverse social attributes. The results of these experiments demonstrate the presence of complex stereotypical behaviors in image generations.
comment: ACL 2023
☆ A Probabilistic Relaxation of the Two-Stage Object Pose Estimation Paradigm
Existing object pose estimation methods commonly require a one-to-one point matching step that forces them to be separated into two consecutive stages: visual correspondence detection (e.g., by matching feature descriptors as part of a perception front-end) followed by geometric alignment (e.g., by optimizing a robust estimation objective for pointcloud registration or perspective-n-point). Instead, we propose a matching-free probabilistic formulation with two main benefits: i) it enables unified and concurrent optimization of both visual correspondence and geometric alignment, and ii) it can represent different plausible modes of the entire distribution of likely poses. This in turn allows for a more graceful treatment of geometric perception scenarios where establishing one-to-one matches between points is conceptually ill-defined, such as textureless, symmetrical and/or occluded objects and scenes where the correct pose is uncertain or there are multiple equally valid solutions.
☆ LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Day
Conversational generative AI has demonstrated remarkable promise for empowering biomedical practitioners, but current investigations focus on unimodal text. Multimodal conversational AI has seen rapid progress by leveraging billions of image-text pairs from the public web, but such general-domain vision-language models still lack sophistication in understanding and conversing about biomedical images. In this paper, we propose a cost-efficient approach for training a vision-language conversational assistant that can answer open-ended research questions of biomedical images. The key idea is to leverage a large-scale, broad-coverage biomedical figure-caption dataset extracted from PubMed Central, use GPT-4 to self-instruct open-ended instruction-following data from the captions, and then fine-tune a large general-domain vision-language model using a novel curriculum learning method. Specifically, the model first learns to align biomedical vocabulary using the figure-caption pairs as is, then learns to master open-ended conversational semantics using GPT-4 generated instruction-following data, broadly mimicking how a layperson gradually acquires biomedical knowledge. This enables us to train a Large Language and Vision Assistant for BioMedicine (LLaVA-Med) in less than 15 hours (with eight A100s). LLaVA-Med exhibits excellent multimodal conversational capability and can follow open-ended instruction to assist with inquiries about a biomedical image. On three standard biomedical visual question answering datasets, LLaVA-Med outperforms previous supervised state-of-the-art on certain metrics. To facilitate biomedical multimodal research, we will release our instruction-following data and the LLaVA-Med model.
comment: 17 pages; Website: https://aka.ms/llava-med
☆ Quantifying Deep Learning Model Uncertainty in Conformal Prediction
Precise estimation of predictive uncertainty in deep neural networks is a critical requirement for reliable decision-making in machine learning and statistical modeling, particularly in the context of medical AI. Conformal Prediction (CP) has emerged as a promising framework for representing the model uncertainty by providing well-calibrated confidence levels for individual predictions. However, the quantification of model uncertainty in conformal prediction remains an active research area, yet to be fully addressed. In this paper, we explore state-of-the-art CP methodologies and their theoretical foundations. We propose a probabilistic approach in quantifying the model uncertainty derived from the produced prediction sets in conformal prediction and provide certified boundaries for the computed uncertainty. By doing so, we allow model uncertainty measured by CP to be compared by other uncertainty quantification methods such as Bayesian (e.g., MC-Dropout and DeepEnsemble) and Evidential approaches.
comment: 7 pages, 4 figures
☆ A Transformer-based representation-learning model with unified processing of multimodal input for clinical diagnostics
During the diagnostic process, clinicians leverage multimodal information, such as chief complaints, medical images, and laboratory-test results. Deep-learning models for aiding diagnosis have yet to meet this requirement. Here we report a Transformer-based representation-learning model as a clinical diagnostic aid that processes multimodal input in a unified manner. Rather than learning modality-specific features, the model uses embedding layers to convert images and unstructured and structured text into visual tokens and text tokens, and bidirectional blocks with intramodal and intermodal attention to learn a holistic representation of radiographs, the unstructured chief complaint and clinical history, structured clinical information such as laboratory-test results and patient demographic information. The unified model outperformed an image-only model and non-unified multimodal diagnosis models in the identification of pulmonary diseases (by 12% and 9%, respectively) and in the prediction of adverse clinical outcomes in patients with COVID-19 (by 29% and 7%, respectively). Leveraging unified multimodal Transformer-based models may help streamline triage of patients and facilitate the clinical decision process.
comment: Accepted by Nature Biomedical Engineering
☆ DeepFake-Adapter: Dual-Level Adapter for DeepFake Detection
Existing deepfake detection methods fail to generalize well to unseen or degraded samples, which can be attributed to the over-fitting of low-level forgery patterns. Here we argue that high-level semantics are also indispensable recipes for generalizable forgery detection. Recently, large pre-trained Vision Transformers (ViTs) have shown promising generalization capability. In this paper, we propose the first parameter-efficient tuning approach for deepfake detection, namely DeepFake-Adapter, to effectively and efficiently adapt the generalizable high-level semantics from large pre-trained ViTs to aid deepfake detection. Given large pre-trained models but limited deepfake data, DeepFake-Adapter introduces lightweight yet dedicated dual-level adapter modules to a ViT while keeping the model backbone frozen. Specifically, to guide the adaptation process to be aware of both global and local forgery cues of deepfake data, 1) we not only insert Globally-aware Bottleneck Adapters in parallel to MLP layers of ViT, 2) but also actively cross-attend Locally-aware Spatial Adapters with features from ViT. Unlike existing deepfake detection methods merely focusing on low-level forgery patterns, the forgery detection process of our model can be regularized by generalizable high-level semantics from a pre-trained ViT and adapted by global and local low-level forgeries of deepfake data. Extensive experiments on several standard deepfake detection benchmarks validate the effectiveness of our approach. Notably, DeepFake-Adapter demonstrates a convincing advantage under cross-dataset and cross-manipulation settings. The source code is released at https://github.com/rshaojimmy/DeepFake-Adapter
comment: Github: https://github.com/rshaojimmy/DeepFake-Adapter
☆ A deep-learning approach to early identification of suggested sexual harassment from videos
Sexual harassment, sexual abuse, and sexual violence are prevalent problems in this day and age. Women's safety is an important issue that needs to be highlighted and addressed. Given this issue, we have studied each of these concerns and the factors that affect it based on images generated from movies. We have classified the three terms (harassment, abuse, and violence) based on the visual attributes present in images depicting these situations. We identified that factors such as facial expression of the victim and perpetrator and unwanted touching had a direct link to identifying the scenes containing sexual harassment, abuse and violence. We also studied and outlined how state-of-the-art explicit content detectors such as Google Cloud Vision API and Clarifai API fail to identify and categorise these images. Based on these definitions and characteristics, we have developed a first-of-its-kind dataset from various Indian movie scenes. These scenes are classified as sexual harassment, sexual abuse, or sexual violence and exported in the PASCAL VOC 1.1 format. Our dataset is annotated on the identified relevant features and can be used to develop and train a deep-learning computer vision model to identify these issues. The dataset is publicly available for research and development.
comment: 7 pages, 4 figures
☆ Spatio-Angular Convolutions for Super-resolution in Diffusion MRI
Diffusion MRI (dMRI) is a widely used imaging modality, but requires long scanning times to acquire high resolution datasets. By leveraging the unique geometry present within this domain, we present a novel approach to dMRI angular super-resolution that extends upon the parametric continuous convolution (PCConv) framework. We introduce several additions to the operation including a Fourier feature mapping, global coordinates, and domain specific context. Using this framework, we build a fully parametric continuous convolution network (PCCNN) and compare against existing models. We demonstrate the PCCNN performs competitively while using significantly less parameters. Moreover, we show that this formulation generalises well to clinically relevant downstream analyses such as fixel-based analysis, and neurite orientation dispersion and density imaging.
☆ Deformable Convolutions and LSTM-based Flexible Event Frame Fusion Network for Motion Deblurring
Event cameras differ from conventional RGB cameras in that they produce asynchronous data sequences. While RGB cameras capture every frame at a fixed rate, event cameras only capture changes in the scene, resulting in sparse and asynchronous data output. Despite the fact that event data carries useful information that can be utilized in motion deblurring of RGB cameras, integrating event and image information remains a challenge. Recent state-of-the-art CNN-based deblurring solutions produce multiple 2-D event frames based on the accumulation of event data over a time period. In most of these techniques, however, the number of event frames is fixed and predefined, which reduces temporal resolution drastically, particularly for scenarios when fast-moving objects are present or when longer exposure times are required. It is also important to note that recent modern cameras (e.g., cameras in mobile phones) dynamically set the exposure time of the image, which presents an additional problem for networks developed for a fixed number of event frames. A Long Short-Term Memory (LSTM)-based event feature extraction module has been developed for addressing these challenges, which enables us to use a dynamically varying number of event frames. Using these modules, we constructed a state-of-the-art deblurring network, Deformable Convolutions and LSTM-based Flexible Event Frame Fusion Network (DLEFNet). It is particularly useful for scenarios in which exposure times vary depending on factors such as lighting conditions or the presence of fast-moving objects in the scene. It has been demonstrated through evaluation results that the proposed method can outperform the existing state-of-the-art networks for deblurring task in synthetic and real-world data sets.
☆ In or Out? Fixing ImageNet Out-of-Distribution Detection Evaluation ICML 2023
Out-of-distribution (OOD) detection is the problem of identifying inputs which are unrelated to the in-distribution task. The OOD detection performance when the in-distribution (ID) is ImageNet-1K is commonly being tested on a small range of test OOD datasets. We find that most of the currently used test OOD datasets, including datasets from the open set recognition (OSR) literature, have severe issues: In some cases more than 50$\%$ of the dataset contains objects belonging to one of the ID classes. These erroneous samples heavily distort the evaluation of OOD detectors. As a solution, we introduce with NINCO a novel test OOD dataset, each sample checked to be ID free, which with its fine-grained range of OOD classes allows for a detailed analysis of an OOD detector's strengths and failure modes, particularly when paired with a number of synthetic "OOD unit-tests". We provide detailed evaluations across a large set of architectures and OOD detection methods on NINCO and the unit-tests, revealing new insights about model weaknesses and the effects of pretraining on OOD detection performance. We provide code and data at https://github.com/j-cb/NINCO.
comment: ICML 2023. Datasets, code and evaluation data at https://github.com/j-cb/NINCO
☆ Geo-Tiles for Semantic Segmentation of Earth Observation Imagery
To cope with the high requirements during the computation of semantic segmentations of earth observation imagery, current state-of-the-art pipelines divide the corresponding data into smaller images. Existing methods and benchmark datasets oftentimes rely on pixel-based tiling schemes or on geo-tiling schemes employed by web mapping applications. The selection of the subimages (comprising size, location and orientation) is crucial since it affects the available context information of each pixel, defines the number of tiles during training, and influences the degree of information degradation while down- and up-sampling the tile contents to the size required by the segmentation model. In this paper we propose a new segmentation pipeline for earth observation imagery relying on a tiling scheme that creates geo-tiles based on the geo-information of the raster data. This approach exhibits several beneficial properties compared to pixel-based or common web mapping approaches. For instance, the proposed tiling scheme shows flexible customization properties regarding tile granularity, tile stride and image boundary alignment, which allows us to perform a tile specific data augmentation during training and a substitution of pixel predictions with limited context information using data of overlapping tiles during inference. Furthermore, the generated tiles show a consistent spatial tile extent w.r.t. heterogeneous sensors, varying recording distances and different latitudes. In our experiments we demonstrate how the proposed tiling system allows to improve the results of current state-of-the-art semantic segmentation models. To foster future research we make the source code publicly available.
☆ Dilated Convolution with Learnable Spacings: beyond bilinear interpolation
Dilated Convolution with Learnable Spacings (DCLS) is a recently proposed variation of the dilated convolution in which the spacings between the non-zero elements in the kernel, or equivalently their positions, are learnable. Non-integer positions are handled via interpolation. Thanks to this trick, positions have well-defined gradients. The original DCLS used bilinear interpolation, and thus only considered the four nearest pixels. Yet here we show that longer range interpolations, and in particular a Gaussian interpolation, allow improving performance on ImageNet1k classification on two state-of-the-art convolutional architectures (ConvNeXt and Conv\-Former), without increasing the number of parameters. The method code is based on PyTorch and is available at https://github.com/K-H-Ismail/Dilated-Convolution-with-Learnable-Spacings-PyTorch
comment: 7 pages
☆ Robust Backdoor Attack with Visible, Semantic, Sample-Specific, and Compatible Triggers
Deep neural networks (DNNs) can be manipulated to exhibit specific behaviors when exposed to specific trigger patterns, without affecting their performance on normal samples. This type of attack is known as a backdoor attack. Recent research has focused on designing invisible triggers for backdoor attacks to ensure visual stealthiness. These triggers have demonstrated strong attack performance even under backdoor defense, which aims to eliminate or suppress the backdoor effect in the model. However, through experimental observations, we have noticed that these carefully designed invisible triggers are often susceptible to visual distortion during inference, such as Gaussian blurring or environmental variations in real-world scenarios. This phenomenon significantly undermines the effectiveness of attacks in practical applications. Unfortunately, this issue has not received sufficient attention and has not been thoroughly investigated. To address this limitation, we propose a novel approach called the Visible, Semantic, Sample-Specific, and Compatible trigger (VSSC-trigger), which leverages a recent powerful image method known as the stable diffusion model. In this approach, a text trigger is utilized as a prompt and combined with a benign image. The resulting combination is then processed by a pre-trained stable diffusion model, generating a corresponding semantic object. This object is seamlessly integrated with the original image, resulting in a new realistic image, referred to as the poisoned image. Extensive experimental results and analysis validate the effectiveness and robustness of our proposed attack method, even in the presence of visual distortion. We believe that the new trigger proposed in this work, along with the proposed idea to address the aforementioned issues, will have significant prospective implications for further advancements in this direction.
☆ UniDiff: Advancing Vision-Language Models with Generative and Discriminative Learning
Recent advances in vision-language pre-training have enabled machines to perform better in multimodal object discrimination (e.g., image-text semantic alignment) and image synthesis (e.g., text-to-image generation). On the other hand, fine-tuning pre-trained models with discriminative or generative capabilities such as CLIP and Stable Diffusion on domain-specific datasets has shown to be effective in various tasks by adapting to specific domains. However, few studies have explored the possibility of learning both discriminative and generative capabilities and leveraging their synergistic effects to create a powerful and personalized multimodal model during fine-tuning. This paper presents UniDiff, a unified multi-modal model that integrates image-text contrastive learning (ITC), text-conditioned image synthesis learning (IS), and reciprocal semantic consistency modeling (RSC). UniDiff effectively learns aligned semantics and mitigates the issue of semantic collapse during fine-tuning on small datasets by leveraging RSC on visual features from CLIP and diffusion models, without altering the pre-trained model's basic architecture. UniDiff demonstrates versatility in both multi-modal understanding and generative tasks. Experimental results on three datasets (Fashion-man, Fashion-woman, and E-commercial Product) showcase substantial enhancements in vision-language retrieval and text-to-image generation, illustrating the advantages of combining discriminative and generative fine-tuning. The proposed UniDiff model establishes a robust pipeline for personalized modeling and serves as a benchmark for future comparisons in the field.
comment: NA
☆ FigGen: Text to Scientific Figure Generation ICLR 2023
The generative modeling landscape has experienced tremendous growth in recent years, particularly in generating natural images and art. Recent techniques have shown impressive potential in creating complex visual compositions while delivering impressive realism and quality. However, state-of-the-art methods have been focusing on the narrow domain of natural images, while other distributions remain unexplored. In this paper, we introduce the problem of text-to-figure generation, that is creating scientific figures of papers from text descriptions. We present FigGen, a diffusion-based approach for text-to-figure as well as the main challenges of the proposed task. Code and models are available at https://github.com/joanrod/figure-diffusion
comment: Published at ICLR 2023 as a Tiny Paper
☆ Learning Across Decentralized Multi-Modal Remote Sensing Archives with Federated Learning
The development of federated learning (FL) methods, which aim to learn from distributed databases (i.e., clients) without accessing data on clients, has recently attracted great attention. Most of these methods assume that the clients are associated with the same data modality. However, remote sensing (RS) images in different clients can be associated with different data modalities that can improve the classification performance when jointly used. To address this problem, in this paper we introduce a novel multi-modal FL framework that aims to learn from decentralized multi-modal RS image archives for RS image classification problems. The proposed framework is made up of three modules: 1) multi-modal fusion (MF); 2) feature whitening (FW); and 3) mutual information maximization (MIM). The MF module performs iterative model averaging to learn without accessing data on clients in the case that clients are associated with different data modalities. The FW module aligns the representations learned among the different clients. The MIM module maximizes the similarity of images from different modalities. Experimental results show the effectiveness of the proposed framework compared to iterative model averaging, which is a widely used algorithm in FL. The code of the proposed framework is publicly available at https://git.tu-berlin.de/rsim/MM-FL.
comment: Accepted at IEEE International Geoscience and Remote Sensing Symposium (IGARSS) 2023. Our code is available at https://git.tu-berlin.de/rsim/MM-FL
☆ FDNeRF: Semantics-Driven Face Reconstruction, Prompt Editing and Relighting with Diffusion Models
The ability to create high-quality 3D faces from a single image has become increasingly important with wide applications in video conferencing, AR/VR, and advanced video editing in movie industries. In this paper, we propose Face Diffusion NeRF (FDNeRF), a new generative method to reconstruct high-quality Face NeRFs from single images, complete with semantic editing and relighting capabilities. FDNeRF utilizes high-resolution 3D GAN inversion and expertly trained 2D latent-diffusion model, allowing users to manipulate and construct Face NeRFs in zero-shot learning without the need for explicit 3D data. With carefully designed illumination and identity preserving loss, as well as multi-modal pre-training, FD-NeRF offers users unparalleled control over the editing process enabling them to create and edit face NeRFs using just single-view images, text prompts, and explicit target lighting. The advanced features of FDNeRF have been designed to produce more impressive results than existing 2D editing approaches that rely on 2D segmentation maps for editable attributes. Experiments show that our FDNeRF achieves exceptionally realistic results and unprecedented flexibility in editing compared with state-of-the-art 3D face reconstruction and editing methods. Our code will be available at https://github.com/BillyXYB/FDNeRF.
☆ Object pop-up: Can we infer 3D objects and their poses from human interactions alone? CVPR'23
The intimate entanglement between objects affordances and human poses is of large interest, among others, for behavioural sciences, cognitive psychology, and Computer Vision communities. In recent years, the latter has developed several object-centric approaches: starting from items, learning pipelines synthesizing human poses and dynamics in a realistic way, satisfying both geometrical and functional expectations. However, the inverse perspective is significantly less explored: Can we infer 3D objects and their poses from human interactions alone? Our investigation follows this direction, showing that a generic 3D human point cloud is enough to pop up an unobserved object, even when the user is just imitating a functionality (e.g., looking through a binocular) without involving a tangible counterpart. We validate our method qualitatively and quantitatively, with synthetic data and sequences acquired for the task, showing applicability for XR/VR. The code is available at https://github.com/ptrvilya/object-popup.
comment: Accepted at CVPR'23
☆ Learning Disentangled Prompts for Compositional Image Synthesis
We study domain-adaptive image synthesis, the problem of teaching pretrained image generative models a new style or concept from as few as one image to synthesize novel images, to better understand the compositional image synthesis. We present a framework that leverages a pretrained class-conditional generation model and visual prompt tuning. Specifically, we propose a novel source class distilled visual prompt that learns disentangled prompts of semantic (e.g., class) and domain (e.g., style) from a few images. Learned domain prompt is then used to synthesize images of any classes in the style of target domain. We conduct studies on various target domains with the number of images ranging from one to a few to many, and show qualitative results which show the compositional generalization of our method. Moreover, we show that our method can help improve zero-shot domain adaptation classification accuracy.
comment: tech report
☆ LiT-4-RSVQA: Lightweight Transformer-based Visual Question Answering in Remote Sensing
Visual question answering (VQA) methods in remote sensing (RS) aim to answer natural language questions with respect to an RS image. Most of the existing methods require a large amount of computational resources, which limits their application in operational scenarios in RS. To address this issue, in this paper we present an effective lightweight transformer-based VQA in RS (LiT-4-RSVQA) architecture for efficient and accurate VQA in RS. Our architecture consists of: i) a lightweight text encoder module; ii) a lightweight image encoder module; iii) a fusion module; and iv) a classification module. The experimental results obtained on a VQA benchmark dataset demonstrate that our proposed LiT-4-RSVQA architecture provides accurate VQA results while significantly reducing the computational requirements on the executing hardware. Our code is publicly available at https://git.tu-berlin.de/rsim/lit4rsvqa.
comment: Accepted at IEEE International Geoscience and Remote Sensing Symposium 2023
☆ Robust T-Loss for Medical Image Segmentation MICCAI 2023
This paper presents a new robust loss function, the T-Loss, for medical image segmentation. The proposed loss is based on the negative log-likelihood of the Student-t distribution and can effectively handle outliers in the data by controlling its sensitivity with a single parameter. This parameter is updated during the backpropagation process, eliminating the need for additional computation or prior information about the level and spread of noisy labels. Our experiments show that the T-Loss outperforms traditional loss functions in terms of dice scores on two public medical datasets for skin lesion and lung segmentation. We also demonstrate the ability of T-Loss to handle different types of simulated label noise, resembling human error. Our results provide strong evidence that the T-Loss is a promising alternative for medical image segmentation where high levels of noise or outliers in the dataset are a typical phenomenon in practice. The project website can be found at https://robust-tloss.github.io
comment: Early accepted to MICCAI 2023
☆ ReFACT: Updating Text-to-Image Models by Editing the Text Encoder
Text-to-image models are trained on extensive amounts of data, leading them to implicitly encode factual knowledge within their parameters. While some facts are useful, others may be incorrect or become outdated (e.g., the current President of the United States). We introduce ReFACT, a novel approach for editing factual knowledge in text-to-image generative models. ReFACT updates the weights of a specific layer in the text encoder, only modifying a tiny portion of the model's parameters, and leaving the rest of the model unaffected. We empirically evaluate ReFACT on an existing benchmark, alongside RoAD, a newly curated dataset. ReFACT achieves superior performance in terms of generalization to related concepts while preserving unrelated concepts. Furthermore, ReFACT maintains image generation quality, making it a valuable tool for updating and correcting factual information in text-to-image models.
☆ Dissecting Arbitrary-scale Super-resolution Capability from Pre-trained Diffusion Generative Models
Diffusion-based Generative Models (DGMs) have achieved unparalleled performance in synthesizing high-quality visual content, opening up the opportunity to improve image super-resolution (SR) tasks. Recent solutions for these tasks often train architecture-specific DGMs from scratch, or require iterative fine-tuning and distillation on pre-trained DGMs, both of which take considerable time and hardware investments. More seriously, since the DGMs are established with a discrete pre-defined upsampling scale, they cannot well match the emerging requirements of arbitrary-scale super-resolution (ASSR), where a unified model adapts to arbitrary upsampling scales, instead of preparing a series of distinct models for each case. These limitations beg an intriguing question: can we identify the ASSR capability of existing pre-trained DGMs without the need for distillation or fine-tuning? In this paper, we take a step towards resolving this matter by proposing Diff-SR, a first ASSR attempt based solely on pre-trained DGMs, without additional training efforts. It is motivated by an exciting finding that a simple methodology, which first injects a specific amount of noise into the low-resolution images before invoking a DGM's backward diffusion process, outperforms current leading solutions. The key insight is determining a suitable amount of noise to inject, i.e., small amounts lead to poor low-level fidelity, while over-large amounts degrade the high-level signature. Through a finely-grained theoretical analysis, we propose the Perceptual Recoverable Field (PRF), a metric that achieves the optimal trade-off between these two factors. Extensive experiments verify the effectiveness, flexibility, and adaptability of Diff-SR, demonstrating superior performance to state-of-the-art solutions under diverse ASSR environments.
☆ DAM-Net: Global Flood Detection from SAR Imagery Using Differential Attention Metric-Based Vision Transformers
The detection of flooded areas using high-resolution synthetic aperture radar (SAR) imagery is a critical task with applications in crisis and disaster management, as well as environmental resource planning. However, the complex nature of SAR images presents a challenge that often leads to an overestimation of the flood extent. To address this issue, we propose a novel differential attention metric-based network (DAM-Net) in this study. The DAM-Net comprises two key components: a weight-sharing Siamese backbone to obtain multi-scale change features of multi-temporal images and tokens containing high-level semantic information of water-body changes, and a temporal differential fusion (TDF) module that integrates semantic tokens and change features to generate flood maps with reduced speckle noise. Specifically, the backbone is split into multiple stages. In each stage, we design three modules, namely, temporal-wise feature extraction (TWFE), cross-temporal change attention (CTCA), and temporal-aware change enhancement (TACE), to effectively extract the change features. In TACE of the last stage, we introduce a class token to record high-level semantic information of water-body changes via the attention mechanism. Another challenge faced by data-driven deep learning algorithms is the limited availability of flood detection datasets. To overcome this, we have created the S1GFloods open-source dataset, a global-scale high-resolution Sentinel-1 SAR image pairs dataset covering 46 global flood events between 2015 and 2022. The experiments on the S1GFloods dataset using the proposed DAM-Net showed top results compared to state-of-the-art methods in terms of overall accuracy, F1-score, and IoU, which reached 97.8%, 96.5%, and 93.2%, respectively. Our dataset and code will be available online at https://github.com/Tamer-Saleh/S1GFlood-Detection.
comment: 16 pages, 11 figures
☆ Prediction of Post-Operative Renal and Pulmonary Complication Using Transformers
Postoperative complications pose a significant challenge in the healthcare industry, resulting in elevated healthcare expenses and prolonged hospital stays, and in rare instances, patient mortality. To improve patient outcomes and reduce healthcare costs, healthcare providers rely on various perioperative risk scores to guide clinical decisions and prioritize care. In recent years, machine learning techniques have shown promise in predicting postoperative complications and fatality, with deep learning models achieving remarkable success in healthcare applications. However, research on the application of deep learning models to intra-operative anesthesia management data is limited. In this paper, we evaluate the performance of transformer-based models in predicting postoperative acute renal failure, postoperative pulmonary complications, and postoperative in-hospital mortality. We compare our method's performance with state-of-the-art tabular data prediction models, including gradient boosting trees and sequential attention models, on a clinical dataset. Our results demonstrate that transformer-based models can achieve superior performance in predicting postoperative complications and outperform traditional machine learning models. This work highlights the potential of deep learning techniques, specifically transformer-based models, in revolutionizing the healthcare industry's approach to postoperative care.
☆ Analyzing the Internals of Neural Radiance Fields
Modern Neural Radiance Fields (NeRFs) learn a mapping from position to volumetric density via proposal network samplers. In contrast to the coarse-to-fine sampling approach with two NeRFs, this offers significant potential for speedups using lower network capacity as the task of mapping spatial coordinates to volumetric density involves no view-dependent effects and is thus much easier to learn. Given that most of the network capacity is utilized to estimate radiance, NeRFs could store valuable density information in their parameters or their deep features. To this end, we take one step back and analyze large, trained ReLU-MLPs used in coarse-to-fine sampling. We find that trained NeRFs, Mip-NeRFs and proposal network samplers map samples with high density to local minima along a ray in activation feature space. We show how these large MLPs can be accelerated by transforming the intermediate activations to a weight estimate, without any modifications to the parameters post-optimization. With our approach, we can reduce the computational requirements of trained NeRFs by up to 50% with only a slight hit in rendering quality and no changes to the training protocol or architecture. We evaluate our approach on a variety of architectures and datasets, showing that our proposition holds in various settings.
comment: project page: nerfinternals.github.io
☆ Can Large Pre-trained Models Help Vision Models on Perception Tasks?
The recent upsurge in pre-trained large models (e.g. GPT-4) has swept across the entire deep learning community. Such powerful large language models (LLMs) demonstrate advanced generative ability and multimodal understanding capability, which quickly achieve new state-of-the-art performances on a variety of benchmarks. The pre-trained LLM usually plays the role as a universal AI model that can conduct various tasks, including context reasoning, article analysis and image content comprehension. However, considering the prohibitively high memory and computational cost for implementing such a large model, the conventional models (such as CNN and ViT), are still essential for many visual perception tasks. In this paper, we propose to enhance the representation ability of ordinary vision models for perception tasks (e.g. image classification) by taking advantage of large pre-trained models. We present a new learning paradigm in which the knowledge extracted from large pre-trained models are utilized to help models like CNN and ViT learn enhanced representations and achieve better performance. Firstly, we curate a high quality description set by prompting a multimodal LLM to generate descriptive text for all training images. Furthermore, we feed these detailed descriptions into a pre-trained encoder to extract text embeddings with rich semantic information that encodes the content of images. During training, text embeddings will serve as extra supervising signals and be aligned with image representations learned by vision models. The alignment process helps vision models learn better and achieve higher accuracy with the assistance of pre-trained LLMs. We conduct extensive experiments to verify that the proposed algorithm consistently improves the performance for various vision models with heterogeneous architectures.
comment: 9 pages, 5 figures
☆ Hyperspectral Target Detection Based on Low-Rank Background Subspace Learning and Graph Laplacian Regularization
Hyperspectral target detection is good at finding dim and small objects based on spectral characteristics. However, existing representation-based methods are hindered by the problem of the unknown background dictionary and insufficient utilization of spatial information. To address these issues, this paper proposes an efficient optimizing approach based on low-rank representation (LRR) and graph Laplacian regularization (GLR). Firstly, to obtain a complete and pure background dictionary, we propose a LRR-based background subspace learning method by jointly mining the low-dimensional structure of all pixels. Secondly, to fully exploit local spatial relationships and capture the underlying geometric structure, a local region-based GLR is employed to estimate the coefficients. Finally, the desired detection map is generated by computing the ratio of representation errors from binary hypothesis testing. The experiments conducted on two benchmark datasets validate the effectiveness and superiority of the approach. For reproduction, the accompanying code is available at https://github.com/shendb2022/LRBSL-GLR.
comment: 4 pages, 3 figures, 1 table
☆ NeuroGF: A Neural Representation for Fast Geodesic Distance and Path Queries
Geodesics are essential in many geometry processing applications. However, traditional algorithms for computing geodesic distances and paths on 3D mesh models are often inefficient and slow. This makes them impractical for scenarios that require extensive querying of arbitrary point-to-point geodesics. Although neural implicit representations have emerged as a popular way of representing 3D shape geometries, there is still no research on representing geodesics with deep implicit functions. To bridge this gap, this paper presents the first attempt to represent geodesics on 3D mesh models using neural implicit functions. Specifically, we introduce neural geodesic fields (NeuroGFs), which are learned to represent the all-pairs geodesics of a given mesh. By using NeuroGFs, we can efficiently and accurately answer queries of arbitrary point-to-point geodesic distances and paths, overcoming the limitations of traditional algorithms. Evaluations on common 3D models show that NeuroGFs exhibit exceptional performance in solving the single-source all-destination (SSAD) and point-to-point geodesics, and achieve high accuracy consistently. Moreover, NeuroGFs offer the unique advantage of encoding both 3D geometry and geodesics in a unified representation. Code is made available at https://github.com/keeganhk/NeuroGF/tree/master.
☆ Universal Test-time Adaptation through Weight Ensembling, Diversity Weighting, and Prior Correction
Since distribution shifts are likely to occur during test-time and can drastically decrease the model's performance, online test-time adaptation (TTA) continues to update the model after deployment, leveraging the current test data. Clearly, a method proposed for online TTA has to perform well for all kinds of environmental conditions. By introducing the variable factors 'domain non-stationarity' and 'temporal correlation', we first unfold all practically relevant settings and define the entity as universal TTA. To tackle the problem of universal TTA, we identify and highlight several challenges a self-training based method has to deal with, including: 1) model bias and the occurrence of trivial solutions when performing entropy minimization on varying sequence lengths with and without multiple domain shifts, 2) loss of generalization which exacerbates the adaptation to future domain shifts and the occurrence of catastrophic forgetting, and 3) performance degradation due to shifts in label prior. To prevent the model from becoming biased, we leverage a dataset and model-agnostic certainty and diversity weighting. In order to maintain generalization and prevent catastrophic forgetting, we propose to continually weight-average the source and adapted model. To compensate for disparities in the label prior during test-time, we propose an adaptive additive prior correction scheme. We evaluate our approach, named ROID, on a wide range of settings, datasets, and models, setting new standards in the field of universal TTA.
☆ Multi-Modal Deep Learning for Multi-Temporal Urban Mapping With a Partly Missing Optical Modality
This paper proposes a novel multi-temporal urban mapping approach using multi-modal satellite data from the Sentinel-1 Synthetic Aperture Radar (SAR) and Sentinel-2 MultiSpectral Instrument (MSI) missions. In particular, it focuses on the problem of a partly missing optical modality due to clouds. The proposed model utilizes two networks to extract features from each modality separately. In addition, a reconstruction network is utilized to approximate the optical features based on the SAR data in case of a missing optical modality. Our experiments on a multi-temporal urban mapping dataset with Sentinel-1 SAR and Sentinel-2 MSI data demonstrate that the proposed method outperforms a multi-modal approach that uses zero values as a replacement for missing optical data, as well as a uni-modal SAR-based approach. Therefore, the proposed method is effective in exploiting multi-modal data, if available, but it also retains its effectiveness in case the optical modality is missing.
comment: 4 pages, 2 figures, accepted for publication in the IGARSS 2023 Proceedings
☆ Wuerstchen: Efficient Pretraining of Text-to-Image Models
We introduce Wuerstchen, a novel technique for text-to-image synthesis that unites competitive performance with unprecedented cost-effectiveness and ease of training on constrained hardware. Building on recent advancements in machine learning, our approach, which utilizes latent diffusion strategies at strong latent image compression rates, significantly reduces the computational burden, typically associated with state-of-the-art models, while preserving, if not enhancing, the quality of generated images. Wuerstchen achieves notable speed improvements at inference time, thereby rendering real-time applications more viable. One of the key advantages of our method lies in its modest training requirements of only 9,200 GPU hours, slashing the usual costs significantly without compromising the end performance. In a comparison against the state-of-the-art, we found the approach to yield strong competitiveness. This paper opens the door to a new line of research that prioritizes both performance and computational accessibility, hence democratizing the use of sophisticated AI technologies. Through Wuerstchen, we demonstrate a compelling stride forward in the realm of text-to-image synthesis, offering an innovative path to explore in future research.
☆ Class Anchor Margin Loss for Content-Based Image Retrieval
The performance of neural networks in content-based image retrieval (CBIR) is highly influenced by the chosen loss (objective) function. The majority of objective functions for neural models can be divided into metric learning and statistical learning. Metric learning approaches require a pair mining strategy that often lacks efficiency, while statistical learning approaches are not generating highly compact features due to their indirect feature optimization. To this end, we propose a novel repeller-attractor loss that falls in the metric learning paradigm, yet directly optimizes for the L2 metric without the need of generating pairs. Our loss is formed of three components. One leading objective ensures that the learned features are attracted to each designated learnable class anchor. The second loss component regulates the anchors and forces them to be separable by a margin, while the third objective ensures that the anchors do not collapse to zero. Furthermore, we develop a more efficient two-stage retrieval system by harnessing the learned class anchors during the first stage of the retrieval process, eliminating the need of comparing the query with every image in the database. We establish a set of four datasets (CIFAR-100, Food-101, SVHN, and Tiny ImageNet) and evaluate the proposed objective in the context of few-shot and full-set training on the CBIR task, by using both convolutional and transformer architectures. Compared to existing objective functions, our empirical evidence shows that the proposed objective is generating superior and more consistent results.
☆ AD-PT: Autonomous Driving Pre-Training with Large-scale Point Cloud Dataset
It is a long-term vision for Autonomous Driving (AD) community that the perception models can learn from a large-scale point cloud dataset, to obtain unified representations that can achieve promising results on different tasks or benchmarks. Previous works mainly focus on the self-supervised pre-training pipeline, meaning that they perform the pre-training and fine-tuning on the same benchmark, which is difficult to attain the performance scalability and cross-dataset application for the pre-training checkpoint. In this paper, for the first time, we are committed to building a large-scale pre-training point-cloud dataset with diverse data distribution, and meanwhile learning generalizable representations from such a diverse pre-training dataset. We formulate the point-cloud pre-training task as a semi-supervised problem, which leverages the few-shot labeled and massive unlabeled point-cloud data to generate the unified backbone representations that can be directly applied to many baseline models and benchmarks, decoupling the AD-related pre-training process and downstream fine-tuning task. During the period of backbone pre-training, by enhancing the scene- and instance-level distribution diversity and exploiting the backbone's ability to learn from unknown instances, we achieve significant performance gains on a series of downstream perception benchmarks including Waymo, nuScenes, and KITTI, under different baseline models like PV-RCNN++, SECOND, CenterPoint.
comment: Code is available at: https://github.com/PJLab-ADG/3DTrans
☆ Rotating Features for Object Discovery
The binding problem in human cognition, concerning how the brain represents and connects objects within a fixed network of neural connections, remains a subject of intense debate. Most machine learning efforts addressing this issue in an unsupervised setting have focused on slot-based methods, which may be limiting due to their discrete nature and difficulty to express uncertainty. Recently, the Complex AutoEncoder was proposed as an alternative that learns continuous and distributed object-centric representations. However, it is only applicable to simple toy data. In this paper, we present Rotating Features, a generalization of complex-valued features to higher dimensions, and a new evaluation procedure for extracting objects from distributed representations. Additionally, we show the applicability of our approach to pre-trained features. Together, these advancements enable us to scale distributed object-centric representations from simple toy to real-world data. We believe this work advances a new paradigm for addressing the binding problem in machine learning and has the potential to inspire further innovation in the field.
☆ Revisit Weakly-Supervised Audio-Visual Video Parsing from the Language Perspective
We focus on the weakly-supervised audio-visual video parsing task (AVVP), which aims to identify and locate all the events in audio/visual modalities. Previous works only concentrate on video-level overall label denoising across modalities, but overlook the segment-level label noise, where adjacent video segments (i.e., 1-second video clips) may contain different events. However, recognizing events in the segment is challenging because its label could be any combination of events that occur in the video. To address this issue, we consider tackling AVVP from the language perspective, since language could freely describe how various events appear in each segment beyond fixed labels. Specifically, we design language prompts to describe all cases of event appearance for each video. Then, the similarity between language prompts and segments is calculated, where the event of the most similar prompt is regarded as the segment-level label. In addition, to deal with the mislabeled segments, we propose to perform dynamic re-weighting on the unreliable segments to adjust their labels. Experiments show that our simple yet effective approach outperforms state-of-the-art methods by a large margin.
☆ FMapping: Factorized Efficient Neural Field Mapping for Real-Time Dense RGB SLAM
In this paper, we introduce FMapping, an efficient neural field mapping framework that facilitates the continuous estimation of a colorized point cloud map in real-time dense RGB SLAM. To achieve this challenging goal without depth, a hurdle is how to improve efficiency and reduce the mapping uncertainty of the RGB SLAM system. To this end, we first build up a theoretical analysis by decomposing the SLAM system into tracking and mapping parts, and the mapping uncertainty is explicitly defined within the frame of neural representations. Based on the analysis, we then propose an effective factorization scheme for scene representation and introduce a sliding window strategy to reduce the uncertainty for scene reconstruction. Specifically, we leverage the factorized neural field to decompose uncertainty into a lower-dimensional space, which enhances robustness to noise and improves training efficiency. We then propose the sliding window sampler to reduce uncertainty by incorporating coherent geometric cues from observed frames during map initialization to enhance convergence. Our factorized neural mapping approach enjoys some advantages, such as low memory consumption, more efficient computation, and fast convergence during map initialization. Experiments on two benchmark datasets show that our method can update the map of high-fidelity colorized point clouds around 2 seconds in real time while requiring no customized CUDA kernels. Additionally, it utilizes x20 fewer parameters than the most concise neural implicit mapping of prior methods for SLAM, e.g., iMAP [ 31] and around x1000 fewer parameters than the state-of-the-art approach, e.g., NICE-SLAM [ 42]. For more details, please refer to our project homepage: https://vlis2022.github.io/fmap/.
☆ MammalNet: A Large-scale Video Benchmark for Mammal Recognition and Behavior Understanding CVPR 2023
Monitoring animal behavior can facilitate conservation efforts by providing key insights into wildlife health, population status, and ecosystem function. Automatic recognition of animals and their behaviors is critical for capitalizing on the large unlabeled datasets generated by modern video devices and for accelerating monitoring efforts at scale. However, the development of automated recognition systems is currently hindered by a lack of appropriately labeled datasets. Existing video datasets 1) do not classify animals according to established biological taxonomies; 2) are too small to facilitate large-scale behavioral studies and are often limited to a single species; and 3) do not feature temporally localized annotations and therefore do not facilitate localization of targeted behaviors within longer video sequences. Thus, we propose MammalNet, a new large-scale animal behavior dataset with taxonomy-guided annotations of mammals and their common behaviors. MammalNet contains over 18K videos totaling 539 hours, which is ~10 times larger than the largest existing animal behavior dataset. It covers 17 orders, 69 families, and 173 mammal categories for animal categorization and captures 12 high-level animal behaviors that received focus in previous animal behavior studies. We establish three benchmarks on MammalNet: standard animal and behavior recognition, compositional low-shot animal and behavior recognition, and behavior detection. Our dataset and code have been made available at: https://mammal-net.github.io.
comment: CVPR 2023 proceeding
☆ We never go out of Style: Motion Disentanglement by Subspace Decomposition of Latent Space CVPR
Real-world objects perform complex motions that involve multiple independent motion components. For example, while talking, a person continuously changes their expressions, head, and body pose. In this work, we propose a novel method to decompose motion in videos by using a pretrained image GAN model. We discover disentangled motion subspaces in the latent space of widely used style-based GAN models that are semantically meaningful and control a single explainable motion component. The proposed method uses only a few $(\approx10)$ ground truth video sequences to obtain such subspaces. We extensively evaluate the disentanglement properties of motion subspaces on face and car datasets, quantitatively and qualitatively. Further, we present results for multiple downstream tasks such as motion editing, and selective motion transfer, e.g. transferring only facial expressions without training for it.
comment: AI for content creation, CVPRW-2023
☆ Unleash the Potential of 3D Point Cloud Modeling with A Calibrated Local Geometry-driven Distance Metric
Quantifying the dissimilarity between two unstructured 3D point clouds is a challenging task, with existing metrics often relying on measuring the distance between corresponding points that can be either inefficient or ineffective. In this paper, we propose a novel distance metric called Calibrated Local Geometry Distance (CLGD), which computes the difference between the underlying 3D surfaces calibrated and induced by a set of reference points. By associating each reference point with two given point clouds through computing its directional distances to them, the difference in directional distances of an identical reference point characterizes the geometric difference between a typical local region of the two point clouds. Finally, CLGD is obtained by averaging the directional distance differences of all reference points. We evaluate CLGD on various optimization and unsupervised learning-based tasks, including shape reconstruction, rigid registration, scene flow estimation, and feature representation. Extensive experiments show that CLGD achieves significantly higher accuracy under all tasks in a memory and computationally efficient manner, compared with existing metrics. As a generic metric, CLGD has the potential to advance 3D point cloud modeling. The source code is publicly available at https://github.com/rsy6318/CLGD.
☆ Label- and slide-free tissue histology using 3D epi-mode quantitative phase imaging and virtual H&E staining
Histological staining of tissue biopsies, especially hematoxylin and eosin (H&E) staining, serves as the benchmark for disease diagnosis and comprehensive clinical assessment of tissue. However, the process is laborious and time-consuming, often limiting its usage in crucial applications such as surgical margin assessment. To address these challenges, we combine an emerging 3D quantitative phase imaging technology, termed quantitative oblique back illumination microscopy (qOBM), with an unsupervised generative adversarial network pipeline to map qOBM phase images of unaltered thick tissues (i.e., label- and slide-free) to virtually stained H&E-like (vH&E) images. We demonstrate that the approach achieves high-fidelity conversions to H&E with subcellular detail using fresh tissue specimens from mouse liver, rat gliosarcoma, and human gliomas. We also show that the framework directly enables additional capabilities such as H&E-like contrast for volumetric imaging. The quality and fidelity of the vH&E images are validated using both a neural network classifier trained on real H&E images and tested on virtual H&E images, and a user study with neuropathologists. Given its simple and low-cost embodiment and ability to provide real-time feedback in vivo, this deep learning-enabled qOBM approach could enable new workflows for histopathology with the potential to significantly save time, labor, and costs in cancer screening, detection, treatment guidance, and more.
comment: 30 pages, 9 main figures, 1 table, 5 supplementary figures
☆ AvatarStudio: Text-driven Editing of 3D Dynamic Human Head Avatars
Capturing and editing full head performances enables the creation of virtual characters with various applications such as extended reality and media production. The past few years witnessed a steep rise in the photorealism of human head avatars. Such avatars can be controlled through different input data modalities, including RGB, audio, depth, IMUs and others. While these data modalities provide effective means of control, they mostly focus on editing the head movements such as the facial expressions, head pose and/or camera viewpoint. In this paper, we propose AvatarStudio, a text-based method for editing the appearance of a dynamic full head avatar. Our approach builds on existing work to capture dynamic performances of human heads using neural radiance field (NeRF) and edits this representation with a text-to-image diffusion model. Specifically, we introduce an optimization strategy for incorporating multiple keyframes representing different camera viewpoints and time stamps of a video performance into a single diffusion model. Using this personalized diffusion model, we edit the dynamic NeRF by introducing view-and-time-aware Score Distillation Sampling (VT-SDS) following a model-based guidance approach. Our method edits the full head in a canonical space, and then propagates these edits to remaining time steps via a pretrained deformation network. We evaluate our method visually and numerically via a user study, and results show that our method outperforms existing approaches. Our experiments validate the design choices of our method and highlight that our edits are genuine, personalized, as well as 3D- and time-consistent.
comment: 17 pages, 17 figures. Project page: https://vcai.mpi-inf.mpg.de/projects/AvatarStudio/
☆ A Novel Driver Distraction Behavior Detection Based on Self-Supervised Learning Framework with Masked Image Modeling
Driver distraction causes a significant number of traffic accidents every year, resulting in economic losses and casualties. Currently, the level of automation in commercial vehicles is far from completely unmanned, and drivers still play an important role in operating and controlling the vehicle. Therefore, driver distraction behavior detection is crucial for road safety. At present, driver distraction detection primarily relies on traditional Convolutional Neural Networks (CNN) and supervised learning methods. However, there are still challenges such as the high cost of labeled datasets, limited ability to capture high-level semantic information, and weak generalization performance. In order to solve these problems, this paper proposes a new self-supervised learning method based on masked image modeling for driver distraction behavior detection. Firstly, a self-supervised learning framework for masked image modeling (MIM) is introduced to solve the serious human and material consumption issues caused by dataset labeling. Secondly, the Swin Transformer is employed as an encoder. Performance is enhanced by reconfiguring the Swin Transformer block and adjusting the distribution of the number of window multi-head self-attention (W-MSA) and shifted window multi-head self-attention (SW-MSA) detection heads across all stages, which leads to model more lightening. Finally, various data augmentation strategies are used along with the best random masking strategy to strengthen the model's recognition and generalization ability. Test results on a large-scale driver distraction behavior dataset show that the self-supervised learning method proposed in this paper achieves an accuracy of 99.60%, approximating the excellent performance of advanced supervised learning methods.
☆ Contrastive Learning MRI Reconstruction
Purpose: We propose a novel contrastive learning latent space representation for MRI datasets with partially acquired scans. We show that this latent space can be utilized for accelerated MR image reconstruction. Theory and Methods: Our novel framework, referred to as COLADA (stands for Contrastive Learning for highly accelerated MR image reconstruction), maximizes the mutual information between differently accelerated images of an MRI scan by using self-supervised contrastive learning. In other words, it attempts to "pull" the latent representations of the same scan together and "push" the latent representations of other scans away. The generated MRI latent space is subsequently utilized for MR image reconstruction and the performance was assessed in comparison to several baseline deep learning reconstruction methods. Furthermore, the quality of the proposed latent space representation was analyzed using Alignment and Uniformity. Results: COLADA comprehensively outperformed other reconstruction methods with robustness to variations in undersampling patterns, pathological abnormalities, and noise in k-space during inference. COLADA proved the high quality of reconstruction on unseen data with minimal fine-tuning. The analysis of representation quality suggests that the contrastive features produced by COLADA are optimally distributed in latent space. Conclusion: To the best of our knowledge, this is the first attempt to utilize contrastive learning on differently accelerated images for MR image reconstruction. The proposed latent space representation has practical usage due to a large number of existing partially sampled datasets. This implies the possibility of exploring self-supervised contrastive learning further to enhance the latent space of MRI for image reconstruction.
☆ Layout and Task Aware Instruction Prompt for Zero-shot Document Image Question Answering
The pre-training-fine-tuning paradigm based on layout-aware multimodal pre-trained models has achieved significant progress on document image question answering. However, domain pre-training and task fine-tuning for additional visual, layout, and task modules prevent them from directly utilizing off-the-shelf instruction-tuning language foundation models, which have recently shown promising potential in zero-shot learning. Contrary to aligning language models to the domain of document image question answering, we align document image question answering to off-the-shell instruction-tuning language foundation models to utilize their zero-shot capability. Specifically, we propose layout and task aware instruction prompt called LATIN-Prompt, which consists of layout-aware document content and task-aware descriptions. The former recovers the layout information among text segments from OCR tools by appropriate spaces and line breaks. The latter ensures that the model generates answers that meet the requirements, especially format requirements, through a detailed description of task. Experimental results on three benchmarks show that LATIN-Prompt can improve the zero-shot performance of instruction-tuning language foundation models on document image question answering and help them achieve comparable levels to SOTAs based on the pre-training-fine-tuning paradigm. Quantitative analysis and qualitative analysis demonstrate the effectiveness of LATIN-Prompt. We provide the code in supplementary and will release the code to facilitate future research.
comment: Code is available at https://github.com/WenjinW/LATIN-Prompt
☆ DiffRoom: Diffusion-based High-Quality 3D Room Reconstruction and Generation
We present DiffRoom, a novel framework for tackling the problem of high-quality 3D indoor room reconstruction and generation, both of which are challenging due to the complexity and diversity of the room geometry. Although diffusion-based generative models have previously demonstrated impressive performance in image generation and object-level 3D generation, they have not yet been applied to room-level 3D generation due to their computationally intensive costs. In DiffRoom, we propose a sparse 3D diffusion network that is efficient and possesses strong generative performance for Truncated Signed Distance Field (TSDF), based on a rough occupancy prior. Inspired by KinectFusion's incremental alignment and fusion of local SDFs, we propose a diffusion-based TSDF fusion approach that iteratively diffuses and fuses TSDFs, facilitating the reconstruction and generation of an entire room environment. Additionally, to ease training, we introduce a curriculum diffusion learning paradigm that speeds up the training convergence process and enables high-quality reconstruction. According to the user study, the mesh quality generated by our DiffRoom can even outperform the ground truth mesh provided by ScanNet.
☆ MEWL: Few-shot multimodal word learning with referential uncertainty ICML 2023
Without explicit feedback, humans can rapidly learn the meaning of words. Children can acquire a new word after just a few passive exposures, a process known as fast mapping. This word learning capability is believed to be the most fundamental building block of multimodal understanding and reasoning. Despite recent advancements in multimodal learning, a systematic and rigorous evaluation is still missing for human-like word learning in machines. To fill in this gap, we introduce the MachinE Word Learning (MEWL) benchmark to assess how machines learn word meaning in grounded visual scenes. MEWL covers human's core cognitive toolkits in word learning: cross-situational reasoning, bootstrapping, and pragmatic learning. Specifically, MEWL is a few-shot benchmark suite consisting of nine tasks for probing various word learning capabilities. These tasks are carefully designed to be aligned with the children's core abilities in word learning and echo the theories in the developmental literature. By evaluating multimodal and unimodal agents' performance with a comparative analysis of human performance, we notice a sharp divergence in human and machine word learning. We further discuss these differences between humans and machines and call for human-like few-shot word learning in machines.
comment: Accepted at ICML 2023
☆ Image generation with shortest path diffusion
The field of image generation has made significant progress thanks to the introduction of Diffusion Models, which learn to progressively reverse a given image corruption. Recently, a few studies introduced alternative ways of corrupting images in Diffusion Models, with an emphasis on blurring. However, these studies are purely empirical and it remains unclear what is the optimal procedure for corrupting an image. In this work, we hypothesize that the optimal procedure minimizes the length of the path taken when corrupting an image towards a given final state. We propose the Fisher metric for the path length, measured in the space of probability distributions. We compute the shortest path according to this metric, and we show that it corresponds to a combination of image sharpening, rather than blurring, and noise deblurring. While the corruption was chosen arbitrarily in previous work, our Shortest Path Diffusion (SPD) determines uniquely the entire spatiotemporal structure of the corruption. We show that SPD improves on strong baselines without any hyperparameter tuning, and outperforms all previous Diffusion Models based on image blurring. Furthermore, any small deviation from the shortest path leads to worse performance, suggesting that SPD provides the optimal procedure to corrupt images. Our work sheds new light on observations made in recent works and provides a new approach to improve diffusion models on images and other types of data.
comment: AD and SF contributed equally
☆ DeSAM: Decoupling Segment Anything Model for Generalizable Medical Image Segmentation
Deep learning based automatic medical image segmentation models often suffer from domain shift, where the models trained on a source domain do not generalize well to other unseen domains. As a vision foundation model with powerful generalization capabilities, Segment Anything Model (SAM) shows potential for improving the cross-domain robustness of medical image segmentation. However, SAM and its fine-tuned models performed significantly worse in fully automatic mode compared to when given manual prompts. Upon further investigation, we discovered that the degradation in performance was related to the coupling effect of poor prompts and mask segmentation. In fully automatic mode, the presence of inevitable poor prompts (such as points outside the mask or boxes significantly larger than the mask) can significantly mislead mask generation. To address the coupling effect, we propose the decoupling SAM (DeSAM). DeSAM modifies SAM's mask decoder to decouple mask generation and prompt embeddings while leveraging pre-trained weights. We conducted experiments on publicly available prostate cross-site datasets. The results show that DeSAM improves dice score by an average of 8.96% (from 70.06% to 79.02%) compared to previous state-of-the-art domain generalization method. Moreover, DeSAM can be trained on personal devices with entry-level GPU since our approach does not rely on tuning the heavyweight image encoder. The code is publicly available at https://github.com/yifangao112/DeSAM.
comment: 12 pages. The code is available at https://github.com/yifangao112/DeSAM
☆ Overcoming Language Bias in Remote Sensing Visual Question Answering via Adversarial Training
The Visual Question Answering (VQA) system offers a user-friendly interface and enables human-computer interaction. However, VQA models commonly face the challenge of language bias, resulting from the learned superficial correlation between questions and answers. To address this issue, in this study, we present a novel framework to reduce the language bias of the VQA for remote sensing data (RSVQA). Specifically, we add an adversarial branch to the original VQA framework. Based on the adversarial branch, we introduce two regularizers to constrain the training process against language bias. Furthermore, to evaluate the performance in terms of language bias, we propose a new metric that combines standard accuracy with the performance drop when incorporating question and random image information. Experimental results demonstrate the effectiveness of our method. We believe that our method can shed light on future work for reducing language bias on the RSVQA task.
☆ Interpretable simultaneous localization of MRI corpus callosum and classification of atypical Parkinsonian disorders using YOLOv5
Structural MRI(S-MRI) is one of the most versatile imaging modality that revolutionized the anatomical study of brain in past decades. The corpus callosum (CC) is the principal white matter fibre tract, enabling all kinds of inter-hemispheric communication. Thus, subtle changes in CC might be associated with various neurological disorders. The present work proposes the potential of YOLOv5-based CC detection framework to differentiate atypical Parkinsonian disorders (PD) from healthy controls (HC). With 3 rounds of hold-out validation, mean classification accuracy of 92% is obtained using the proposed method on a proprietary dataset consisting of 20 healthy subjects and 20 cases of APDs, with an improvement of 5% over SOTA methods (CC morphometry and visual texture analysis) that used the same dataset. Subsequently, in order to incorporate the explainability of YOLO predictions, Eigen CAM based heatmap is generated for identifying the most important sub-region in CC that leads to the classification. The result of Eigen CAM showed CC mid-body as the most distinguishable sub-region in classifying APDs and HC, which is in-line with SOTA methodologies and the current prevalent understanding in medicine.
☆ MindBigData 2023 MNIST-8B The 8 billion datapoints Multimodal Dataset of Brain Signals
MindBigData 2023 MNIST-8B is the largest, to date (June 1st 2023), brain signals open dataset created for Machine Learning, based on EEG signals from a single subject captured using a custom 128 channels device, replicating the full 70,000 digits from Yaan LeCun et all MNIST dataset. The brain signals were captured while the subject was watching the pixels of the original digits one by one on a screen and listening at the same time to the spoken number 0 to 9 from the real label. The data, collection procedures, hardware and software created are described in detail, background extra information and other related datasets can be found at our previous paper MindBigData 2022: A Large Dataset of Brain Signals.
comment: 9 pages, 10 figures
☆ S$^2$ME: Spatial-Spectral Mutual Teaching and Ensemble Learning for Scribble-supervised Polyp Segmentation MICCAI 2023
Fully-supervised polyp segmentation has accomplished significant triumphs over the years in advancing the early diagnosis of colorectal cancer. However, label-efficient solutions from weak supervision like scribbles are rarely explored yet primarily meaningful and demanding in medical practice due to the expensiveness and scarcity of densely-annotated polyp data. Besides, various deployment issues, including data shifts and corruption, put forward further requests for model generalization and robustness. To address these concerns, we design a framework of Spatial-Spectral Dual-branch Mutual Teaching and Entropy-guided Pseudo Label Ensemble Learning (S$^2$ME). Concretely, for the first time in weakly-supervised medical image segmentation, we promote the dual-branch co-teaching framework by leveraging the intrinsic complementarity of features extracted from the spatial and spectral domains and encouraging cross-space consistency through collaborative optimization. Furthermore, to produce reliable mixed pseudo labels, which enhance the effectiveness of ensemble learning, we introduce a novel adaptive pixel-wise fusion technique based on the entropy guidance from the spatial and spectral branches. Our strategy efficiently mitigates the deleterious effects of uncertainty and noise present in pseudo labels and surpasses previous alternatives in terms of efficacy. Ultimately, we formulate a holistic optimization objective to learn from the hybrid supervision of scribbles and pseudo labels. Extensive experiments and evaluation on four public datasets demonstrate the superiority of our method regarding in-distribution accuracy, out-of-distribution generalization, and robustness, highlighting its promising clinical significance. Our code is available at https://github.com/lofrienger/S2ME.
comment: MICCAI 2023 Early Acceptance
☆ Exploring Open-Vocabulary Semantic Segmentation without Human Labels
Semantic segmentation is a crucial task in computer vision that involves segmenting images into semantically meaningful regions at the pixel level. However, existing approaches often rely on expensive human annotations as supervision for model training, limiting their scalability to large, unlabeled datasets. To address this challenge, we present ZeroSeg, a novel method that leverages the existing pretrained vision-language (VL) model (e.g. CLIP) to train open-vocabulary zero-shot semantic segmentation models. Although acquired extensive knowledge of visual concepts, it is non-trivial to exploit knowledge from these VL models to the task of semantic segmentation, as they are usually trained at an image level. ZeroSeg overcomes this by distilling the visual concepts learned by VL models into a set of segment tokens, each summarizing a localized region of the target image. We evaluate ZeroSeg on multiple popular segmentation benchmarks, including PASCAL VOC 2012, PASCAL Context, and COCO, in a zero-shot manner (i.e., no training or adaption on target segmentation datasets). Our approach achieves state-of-the-art performance when compared to other zero-shot segmentation methods under the same training data, while also performing competitively compared to strongly supervised methods. Finally, we also demonstrated the effectiveness of ZeroSeg on open-vocabulary segmentation, through both human studies and qualitative visualizations.
☆ Evaluation of Multi-indicator And Multi-organ Medical Image Segmentation Models
In recent years, "U-shaped" neural networks featuring encoder and decoder structures have gained popularity in the field of medical image segmentation. Various variants of this model have been developed. Nevertheless, the evaluation of these models has received less attention compared to model development. In response, we propose a comprehensive method for evaluating medical image segmentation models for multi-indicator and multi-organ (named MIMO). MIMO allows models to generate independent thresholds which are then combined with multi-indicator evaluation and confidence estimation to screen and measure each organ. As a result, MIMO offers detailed information on the segmentation of each organ in each sample, thereby aiding developers in analyzing and improving the model. Additionally, MIMO can produce concise usability and comprehensiveness scores for different models. Models with higher scores are deemed to be excellent models, which is convenient for clinical evaluation. Our research tests eight different medical image segmentation models on two abdominal multi-organ datasets and evaluates them from four perspectives: correctness, confidence estimation, Usable Region and MIMO. Furthermore, robustness experiments are tested. Experimental results demonstrate that MIMO offers novel insights into multi-indicator and multi-organ medical image evaluation and provides a specific and concise measure for the usability and comprehensiveness of the model. Code: https://github.com/SCUT-ML-GUO/MIMO
☆ Edge-guided Representation Learning for Underwater Object Detection
Underwater object detection (UOD) is crucial for marine economic development, environmental protection, and the planet's sustainable development. The main challenges of this task arise from low-contrast, small objects, and mimicry of aquatic organisms. The key to addressing these challenges is to focus the model on obtaining more discriminative information. We observe that the edges of underwater objects are highly unique and can be distinguished from low-contrast or mimicry environments based on their edges. Motivated by this observation, we propose an Edge-guided Representation Learning Network, termed ERL-Net, that aims to achieve discriminative representation learning and aggregation under the guidance of edge cues. Firstly, we introduce an edge-guided attention module to model the explicit boundary information, which generates more discriminative features. Secondly, a feature aggregation module is proposed to aggregate the multi-scale discriminative features by regrouping them into three levels, effectively aggregating global and local information for locating and recognizing underwater objects. Finally, we propose a wide and asymmetric receptive field block to enable features to have a wider receptive field, allowing the model to focus on more small object information. Comprehensive experiments on three challenging underwater datasets show that our method achieves superior performance on the UOD task.
☆ Out-of-distribution forgetting: vulnerability of continual learning to intra-class distribution shift
Continual learning (CL) is an important technique to allow artificial neural networks to work in open environments. CL enables a system to learn new tasks without severe interference to its performance on old tasks, i.e., overcome the problems of catastrophic forgetting. In joint learning, it is well known that the out-of-distribution (OOD) problem caused by intentional attacks or environmental perturbations will severely impair the ability of networks to generalize. In this work, we reported a special form of catastrophic forgetting raised by the OOD problem in continual learning settings, and we named it out-of-distribution forgetting (OODF). In continual image classification tasks, we found that for a given category, introducing an intra-class distribution shift significantly impaired the recognition accuracy of CL methods for that category during subsequent learning. Interestingly, this phenomenon is special for CL as the same level of distribution shift had only negligible effects in the joint learning scenario. We verified that CL methods without dedicating subnetworks for individual tasks are all vulnerable to OODF. Moreover, OODF does not depend on any specific way of shifting the distribution, suggesting it is a risk for CL in a wide range of circumstances. Taken together, our work identified an under-attended risk during CL, highlighting the importance of developing approaches that can overcome OODF.
☆ End-to-end Knowledge Retrieval with Multi-modal Queries ACL 2023
We investigate knowledge retrieval with multi-modal queries, i.e. queries containing information split across image and text inputs, a challenging task that differs from previous work on cross-modal retrieval. We curate a new dataset called ReMuQ for benchmarking progress on this task. ReMuQ requires a system to retrieve knowledge from a large corpus by integrating contents from both text and image queries. We introduce a retriever model ``ReViz'' that can directly process input text and images to retrieve relevant knowledge in an end-to-end fashion without being dependent on intermediate modules such as object detectors or caption generators. We introduce a new pretraining task that is effective for learning knowledge retrieval with multimodal queries and also improves performance on downstream tasks. We demonstrate superior performance in retrieval on two datasets (ReMuQ and OK-VQA) under zero-shot settings as well as further improvements when finetuned on these datasets.
comment: ACL 2023
☆ Introduction to Medical Imaging Informatics
Medical imaging informatics is a rapidly growing field that combines the principles of medical imaging and informatics to improve the acquisition, management, and interpretation of medical images. This chapter introduces the basic concepts of medical imaging informatics, including image processing, feature engineering, and machine learning. It also discusses the recent advancements in computer vision and deep learning technologies and how they are used to develop new quantitative image markers and prediction models for disease detection, diagnosis, and prognosis prediction. By covering the basic knowledge of medical imaging informatics, this chapter provides a foundation for understanding the role of informatics in medicine and its potential impact on patient care.
comment: 17 pages, 11 figures, 2 tables; Acceptance of the chapter for the Springer book "Data-driven approaches to medical imaging"
☆ Controllable Motion Diffusion Model
Generating realistic and controllable motions for virtual characters is a challenging task in computer animation, and its implications extend to games, simulations, and virtual reality. Recent studies have drawn inspiration from the success of diffusion models in image generation, demonstrating the potential for addressing this task. However, the majority of these studies have been limited to offline applications that target at sequence-level generation that generates all steps simultaneously. To enable real-time motion synthesis with diffusion models in response to time-varying control signals, we propose the framework of the Controllable Motion Diffusion Model (COMODO). Our framework begins with an auto-regressive motion diffusion model (A-MDM), which generates motion sequences step by step. In this way, simply using the standard DDPM algorithm without any additional complexity, our framework is able to generate high-fidelity motion sequences over extended periods with different types of control signals. Then, we propose our reinforcement learning-based controller and controlling strategies on top of the A-MDM model, so that our framework can steer the motion synthesis process across multiple tasks, including target reaching, joystick-based control, goal-oriented control, and trajectory following. The proposed framework enables the real-time generation of diverse motions that react adaptively to user commands on-the-fly, thereby enhancing the overall user experience. Besides, it is compatible with the inpainting-based editing methods and can predict much more diverse motions without additional fine-tuning of the basic motion generation models. We conduct comprehensive experiments to evaluate the effectiveness of our framework in performing various tasks and compare its performance against state-of-the-art methods.
☆ Adapting Pre-trained Language Models to Vision-Language Tasks via Dynamic Visual Prompting
Pre-trained language models (PLMs) have played an increasing role in multimedia research. In terms of vision-language (VL) tasks, they often serve as a language encoder and still require an additional fusion network for VL reasoning, resulting in excessive memory overhead. In this paper, we focus on exploring PLMs as a stand-alone model for VL reasoning tasks. Inspired by the recently popular prompt tuning, we first prove that the processed visual features can be also projected onto the semantic space of PLMs and act as prompt tokens to bridge the gap between single- and multi-modal learning. However, this solution exhibits obvious redundancy in visual information and model inference, and the placement of prompt tokens also greatly affects the final performance. Based on these observations, we further propose a novel transfer learning approach for PLMs, termed Dynamic Visual Prompting (DVP). Concretely, DVP first deploys a cross-attention module to obtain text-related and compact visual prompt tokens, thereby greatly reducing the input length of PLMs. To obtain the optimal placement, we also equip DVP with a reinforcement-learning based search algorithm, which can automatically merge DVP with PLMs for different VL tasks via a very short search process. In addition, we also experiment DVP with the recently popular adapter approach to keep the most parameters of PLMs intact when adapting to VL tasks, helping PLMs achieve a quick shift between single- and multi-modal tasks. We apply DVP to two representative PLMs, namely BERT and T5, and conduct extensive experiments on a set of VL reasoning benchmarks including VQA2.0, GQA and SNLIVE. The experimental results not only show the advantage of DVP on efficiency and performance, but also confirm its superiority in adapting pre-trained language models to VL tasks.
☆ Towards Interactive Image Inpainting via Sketch Refinement
One tough problem of image inpainting is to restore complex structures in the corrupted regions. It motivates interactive image inpainting which leverages additional hints, e.g., sketches, to assist the inpainting process. Sketch is simple and intuitive to end users, but meanwhile has free forms with much randomness. Such randomness may confuse the inpainting models, and incur severe artifacts in completed images. To address this problem, we propose a two-stage image inpainting method termed SketchRefiner. In the first stage, we propose using a cross-correlation loss function to robustly calibrate and refine the user-provided sketches in a coarse-to-fine fashion. In the second stage, we learn to extract informative features from the abstracted sketches in the feature space and modulate the inpainting process. We also propose an algorithm to simulate real sketches automatically and build a test protocol with different applications. Experimental results on public datasets demonstrate that SketchRefiner effectively utilizes sketch information and eliminates the artifacts due to the free-form sketches. Our method consistently outperforms the state-of-the-art ones both qualitatively and quantitatively, meanwhile revealing great potential in real-world applications. Our code and dataset are available.
☆ Discriminative Deep Feature Visualization for Explainable Face Recognition
Despite the huge success of deep convolutional neural networks in face recognition (FR) tasks, current methods lack explainability for their predictions because of their "black-box" nature. In recent years, studies have been carried out to give an interpretation of the decision of a deep FR system. However, the affinity between the input facial image and the extracted deep features has not been explored. This paper contributes to the problem of explainable face recognition by first conceiving a face reconstruction-based explanation module, which reveals the correspondence between the deep feature and the facial regions. To further interpret the decision of an FR model, a novel visual saliency explanation algorithm has been proposed. It provides insightful explanation by producing visual saliency maps that represent similar and dissimilar regions between input faces. A detailed analysis has been presented for the generated visual explanation to show the effectiveness of the proposed method.
☆ Lightweight Vision Transformer with Bidirectional Interaction
Recent advancements in vision backbones have significantly improved their performance by simultaneously modeling images' local and global contexts. However, the bidirectional interaction between these two contexts has not been well explored and exploited, which is important in the human visual system. This paper proposes a Fully Adaptive Self-Attention (FASA) mechanism for vision transformer to model the local and global information as well as the bidirectional interaction between them in context-aware ways. Specifically, FASA employs self-modulated convolutions to adaptively extract local representation while utilizing self-attention in down-sampled space to extract global representation. Subsequently, it conducts a bidirectional adaptation process between local and global representation to model their interaction. In addition, we introduce a fine-grained downsampling strategy to enhance the down-sampled self-attention mechanism for finer-grained global perception capability. Based on FASA, we develop a family of lightweight vision backbones, Fully Adaptive Transformer (FAT) family. Extensive experiments on multiple vision tasks demonstrate that FAT achieves impressive performance. Notably, FAT accomplishes a 77.6% accuracy on ImageNet-1K using only 4.5M parameters and 0.7G FLOPs, which surpasses the most advanced ConvNets and Transformers with similar model size and computational costs. Moreover, our model exhibits faster speed on modern GPU compared to other models. Code will be available at https://github.com/qhfan/FAT.
☆ Teacher Agent: A Non-Knowledge Distillation Method for Rehearsal-based Video Incremental Learning
With the rise in popularity of video-based social media, new categories of videos are constantly being generated, creating an urgent need for robust incremental learning techniques for video understanding. One of the biggest challenges in this task is catastrophic forgetting, where the network tends to forget previously learned data while learning new categories. To overcome this issue, knowledge distillation is a widely used technique for rehearsal-based video incremental learning that involves transferring important information on similarities among different categories to enhance the student model. Therefore, it is preferable to have a strong teacher model to guide the students. However, the limited performance of the network itself and the occurrence of catastrophic forgetting can result in the teacher network making inaccurate predictions for some memory exemplars, ultimately limiting the student network's performance. Based on these observations, we propose a teacher agent capable of generating stable and accurate soft labels to replace the output of the teacher model. This method circumvents the problem of knowledge misleading caused by inaccurate predictions of the teacher model and avoids the computational overhead of loading the teacher model for knowledge distillation. Extensive experiments demonstrate the advantages of our method, yielding significant performance improvements while utilizing only half the resolution of video clips in the incremental phases as input compared to recent state-of-the-art methods. Moreover, our method surpasses the performance of joint training when employing four times the number of samples in episodic memory.
comment: Under review; Do We Really Need Knowledge Distillation for Class-incremental Video Learning?
☆ Symmetric Uncertainty-Aware Feature Transmission for Depth Super-Resolution ACM MM 22
Color-guided depth super-resolution (DSR) is an encouraging paradigm that enhances a low-resolution (LR) depth map guided by an extra high-resolution (HR) RGB image from the same scene. Existing methods usually use interpolation to upscale the depth maps before feeding them into the network and transfer the high-frequency information extracted from HR RGB images to guide the reconstruction of depth maps. However, the extracted high-frequency information usually contains textures that are not present in depth maps in the existence of the cross-modality gap, and the noises would be further aggravated by interpolation due to the resolution gap between the RGB and depth images. To tackle these challenges, we propose a novel Symmetric Uncertainty-aware Feature Transmission (SUFT) for color-guided DSR. (1) For the resolution gap, SUFT builds an iterative up-and-down sampling pipeline, which makes depth features and RGB features spatially consistent while suppressing noise amplification and blurring by replacing common interpolated pre-upsampling. (2) For the cross-modality gap, we propose a novel Symmetric Uncertainty scheme to remove parts of RGB information harmful to the recovery of HR depth maps. Extensive experiments on benchmark datasets and challenging real-world settings suggest that our method achieves superior performance compared to state-of-the-art methods. Our code and models are available at https://github.com/ShiWuxuan/SUFT.
comment: 10 pages, 9 figures, accepted by the 30th ACM International Conference on Multimedia (ACM MM 22)
☆ HySpecNet-11k: A Large-Scale Hyperspectral Dataset for Benchmarking Learning-Based Hyperspectral Image Compression Methods
The development of learning-based hyperspectral image compression methods has recently attracted great attention in remote sensing. Such methods require a high number of hyperspectral images to be used during training to optimize all parameters and reach a high compression performance. However, existing hyperspectral datasets are not sufficient to train and evaluate learning-based compression methods, which hinders the research in this field. To address this problem, in this paper we present HySpecNet-11k that is a large-scale hyperspectral benchmark dataset made up of 11,483 nonoverlapping image patches. Each patch is a portion of 128 $\times$ 128 pixels with 224 spectral bands and a ground sample distance of 30 m. We exploit HySpecNet-11k to benchmark the current state of the art in learning-based hyperspectral image compression by focussing our attention on various 1D, 2D and 3D convolutional autoencoder architectures. Nevertheless, HySpecNet-11k can be used for any unsupervised learning task in the framework of hyperspectral image analysis. The dataset, our code and the pre-trained weights are publicly available at https://hyspecnet.rsim.berlin.
☆ Large Scale Generative Multimodal Attribute Extraction for E-commerce Attributes ACL 2023
E-commerce websites (e.g. Amazon) have a plethora of structured and unstructured information (text and images) present on the product pages. Sellers often either don't label or mislabel values of the attributes (e.g. color, size etc.) for their products. Automatically identifying these attribute values from an eCommerce product page that contains both text and images is a challenging task, especially when the attribute value is not explicitly mentioned in the catalog. In this paper, we present a scalable solution for this problem where we pose attribute extraction problem as a question-answering task, which we solve using \textbf{MXT}, consisting of three key components: (i) \textbf{M}AG (Multimodal Adaptation Gate), (ii) \textbf{X}ception network, and (iii) \textbf{T}5 encoder-decoder. Our system consists of a generative model that \emph{generates} attribute-values for a given product by using both textual and visual characteristics (e.g. images) of the product. We show that our system is capable of handling zero-shot attribute prediction (when attribute value is not seen in training data) and value-absent prediction (when attribute value is not mentioned in the text) which are missing in traditional classification-based and NER-based models respectively. We have trained our models using distant supervision, removing dependency on human labeling, thus making them practical for real-world applications. With this framework, we are able to train a single model for 1000s of (product-type, attribute) pairs, thus reducing the overhead of training and maintaining separate models. Extensive experiments on two real world datasets show that our framework improves the absolute recall@90P by 10.16\% and 6.9\% from the existing state of the art models. In a popular e-commerce store, we have deployed our models for 1000s of (product-type, attribute) pairs.
comment: ACL 2023 Industry Track, 8 Pages
☆ Example-based Motion Synthesis via Generative Motion Matching SIGGRAPH 2023
We present GenMM, a generative model that "mines" as many diverse motions as possible from a single or few example sequences. In stark contrast to existing data-driven methods, which typically require long offline training time, are prone to visual artifacts, and tend to fail on large and complex skeletons, GenMM inherits the training-free nature and the superior quality of the well-known Motion Matching method. GenMM can synthesize a high-quality motion within a fraction of a second, even with highly complex and large skeletal structures. At the heart of our generative framework lies the generative motion matching module, which utilizes the bidirectional visual similarity as a generative cost function to motion matching, and operates in a multi-stage framework to progressively refine a random guess using exemplar motion matches. In addition to diverse motion generation, we show the versatility of our generative framework by extending it to a number of scenarios that are not possible with motion matching alone, including motion completion, key frame-guided generation, infinite looping, and motion reassembly. Code and data for this paper are at https://wyysf-98.github.io/GenMM/
comment: SIGGRAPH 2023. Project page: https://wyysf-98.github.io/GenMM/, Video: https://www.youtube.com/watch?v=lehnxcade4I
☆ Graph Switching Dynamical Systems ICML 2023
Dynamical systems with complex behaviours, e.g. immune system cells interacting with a pathogen, are commonly modelled by splitting the behaviour into different regimes, or modes, each with simpler dynamics, and then learning the switching behaviour from one mode to another. Switching Dynamical Systems (SDS) are a powerful tool that automatically discovers these modes and mode-switching behaviour from time series data. While effective, these methods focus on independent objects, where the modes of one object are independent of the modes of the other objects. In this paper, we focus on the more general interacting object setting for switching dynamical systems, where the per-object dynamics also depends on an unknown and dynamically changing subset of other objects and their modes. To this end, we propose a novel graph-based approach for switching dynamical systems, GRAph Switching dynamical Systems (GRASS), in which we use a dynamic graph to characterize interactions between objects and learn both intra-object and inter-object mode-switching behaviour. We introduce two new datasets for this setting, a synthesized ODE-driven particles dataset and a real-world Salsa Couple Dancing dataset. Experiments show that GRASS can consistently outperforms previous state-of-the-art methods.
comment: ICML 2023
☆ How Do ConvNets Understand Image Intensity?
Convolutional Neural Networks (ConvNets) usually rely on edge/shape information to classify images. Visualization methods developed over the last decade confirm that ConvNets rely on edge information. We investigate situations where the ConvNet needs to rely on image intensity in addition to shape. We show that the ConvNet relies on image intensity information using visualization.
☆ Addressing Negative Transfer in Diffusion Models
Diffusion-based generative models have achieved remarkable success in various domains. It trains a model on denoising tasks that encompass different noise levels simultaneously, representing a form of multi-task learning (MTL). However, analyzing and improving diffusion models from an MTL perspective remains under-explored. In particular, MTL can sometimes lead to the well-known phenomenon of $\textit{negative transfer}$, which results in the performance degradation of certain tasks due to conflicts between tasks. In this paper, we aim to analyze diffusion training from an MTL standpoint, presenting two key observations: $\textbf{(O1)}$ the task affinity between denoising tasks diminishes as the gap between noise levels widens, and $\textbf{(O2)}$ negative transfer can arise even in the context of diffusion training. Building upon these observations, our objective is to enhance diffusion training by mitigating negative transfer. To achieve this, we propose leveraging existing MTL methods, but the presence of a huge number of denoising tasks makes this computationally expensive to calculate the necessary per-task loss or gradient. To address this challenge, we propose clustering the denoising tasks into small task clusters and applying MTL methods to them. Specifically, based on $\textbf{(O2)}$, we employ interval clustering to enforce temporal proximity among denoising tasks within clusters. We show that interval clustering can be solved with dynamic programming and utilize signal-to-noise ratio, timestep, and task affinity for clustering objectives. Through this, our approach addresses the issue of negative transfer in diffusion models by allowing for efficient computation of MTL methods. We validate the proposed clustering and its integration with MTL methods through various experiments, demonstrating improved sample quality of diffusion models.
comment: 22 pages, 12 figures, under review
☆ CALICO: Self-Supervised Camera-LiDAR Contrastive Pre-training for BEV Perception
Perception is crucial in the realm of autonomous driving systems, where bird's eye view (BEV)-based architectures have recently reached state-of-the-art performance. The desirability of self-supervised representation learning stems from the expensive and laborious process of annotating 2D and 3D data. Although previous research has investigated pretraining methods for both LiDAR and camera-based 3D object detection, a unified pretraining framework for multimodal BEV perception is missing. In this study, we introduce CALICO, a novel framework that applies contrastive objectives to both LiDAR and camera backbones. Specifically, CALICO incorporates two stages: point-region contrast (PRC) and region-aware distillation (RAD). PRC better balances the region- and scene-level representation learning on the LiDAR modality and offers significant performance improvement compared to existing methods. RAD effectively achieves contrastive distillation on our self-trained teacher model. CALICO's efficacy is substantiated by extensive evaluations on 3D object detection and BEV map segmentation tasks, where it delivers significant performance improvements. Notably, CALICO outperforms the baseline method by 10.5% and 8.6% on NDS and mAP. Moreover, CALICO boosts the robustness of multimodal 3D object detection against adversarial attacks and corruption. Additionally, our framework can be tailored to different backbones and heads, positioning it as a promising approach for multimodal BEV perception.
☆ SelFLoc: Selective Feature Fusion for Large-scale Point Cloud-based Place Recognition
Point cloud-based place recognition is crucial for mobile robots and autonomous vehicles, especially when the global positioning sensor is not accessible. LiDAR points are scattered on the surface of objects and buildings, which have strong shape priors along different axes. To enhance message passing along particular axes, Stacked Asymmetric Convolution Block (SACB) is designed, which is one of the main contributions in this paper. Comprehensive experiments demonstrate that asymmetric convolution and its corresponding strategies employed by SACB can contribute to the more effective representation of point cloud feature. On this basis, Selective Feature Fusion Block (SFFB), which is formed by stacking point- and channel-wise gating layers in a predefined sequence, is proposed to selectively boost salient local features in certain key regions, as well as to align the features before fusion phase. SACBs and SFFBs are combined to construct a robust and accurate architecture for point cloud-based place recognition, which is termed SelFLoc. Comparative experimental results show that SelFLoc achieves the state-of-the-art (SOTA) performance on the Oxford and other three in-house benchmarks with an improvement of 1.6 absolute percentages on mean average recall@1.
☆ Confidence Intervals for Error Rates in Matching Tasks: Critical Review and Recommendations
Matching algorithms are commonly used to predict matches between items in a collection. For example, in 1:1 face verification, a matching algorithm predicts whether two face images depict the same person. Accurately assessing the uncertainty of the error rates of such algorithms can be challenging when data are dependent and error rates are low, two aspects that have been often overlooked in the literature. In this work, we review methods for constructing confidence intervals for error rates in matching tasks such as 1:1 face verification. We derive and examine the statistical properties of these methods and demonstrate how coverage and interval width vary with sample size, error rates, and degree of data dependence using both synthetic and real-world datasets. Based on our findings, we provide recommendations for best practices for constructing confidence intervals for error rates in matching tasks.
comment: 32 pages, 8 figures
☆ Consistency-guided Prompt Learning for Vision-Language Models
We propose Consistency-guided Prompt learning (CoPrompt), a new fine-tuning method for vision-language models that addresses the challenge of improving the generalization capability of large foundation models while fine-tuning them on downstream tasks in a few-shot setting. The basic idea of CoPrompt is to enforce a consistency constraint in the prediction of the trainable and pre-trained models to prevent overfitting on the downstream task. Additionally, we introduce the following two components into our consistency constraint to further boost the performance: enforcing consistency on two perturbed inputs and combining two dominant paradigms of tuning, prompting and adapter. Enforcing consistency on perturbed input further regularizes the consistency constraint, effectively improving generalization, while tuning additional parameters with prompting and adapters improves the performance on downstream tasks. Extensive experiments show that CoPrompt outperforms existing methods on a range of evaluation suites, including base-to-novel generalization, domain generalization, and cross-dataset evaluation tasks. On the generalization task, CoPrompt improves the state-of-the-art by 2.09% on the zero-shot task and 1.93% on the harmonic mean over 11 recognition datasets. Detailed ablation studies show the effectiveness of each of the components in CoPrompt.
☆ Identifying Visible Tissue in Intraoperative Ultrasound Images during Brain Surgery: A Method and Application
Intraoperative ultrasound scanning is a demanding visuotactile task. It requires operators to simultaneously localise the ultrasound perspective and manually perform slight adjustments to the pose of the probe, making sure not to apply excessive force or breaking contact with the tissue, whilst also characterising the visible tissue. In this paper, we propose a method for the identification of the visible tissue, which enables the analysis of ultrasound probe and tissue contact via the detection of acoustic shadow and construction of confidence maps of the perceptual salience. Detailed validation with both in vivo and phantom data is performed. First, we show that our technique is capable of achieving state of the art acoustic shadow scan line classification - with an average binary classification accuracy on unseen data of 0.87. Second, we show that our framework for constructing confidence maps is able to produce an ideal response to a probe's pose that is being oriented in and out of optimality - achieving an average RMSE across five scans of 0.174. The performance evaluation justifies the potential clinical value of the method which can be used both to assist clinical training and optimise robot-assisted ultrasound tissue scanning.
comment: IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
☆ Event-based Visual Odometry with Full Temporal Resolution via Continuous-time Gaussian Process Regression
Event-based cameras asynchronously capture individual visual changes in a scene. This makes them more robust than traditional frame-based cameras to highly dynamic motions and poor illumination. It also means that every measurement in a scene can occur at a unique time. Handling these different measurement times is a major challenge of using event-based cameras. It is often addressed in visual odometry (VO) pipelines by approximating temporally close measurements as occurring at one common time. This grouping simplifies the estimation problem but sacrifices the inherent temporal resolution of event-based cameras. This paper instead presents a complete stereo VO pipeline that estimates directly with individual event-measurement times without requiring any grouping or approximation. It uses continuous-time trajectory estimation to maintain the temporal fidelity and asynchronous nature of event-based cameras through Gaussian process regression with a physically motivated prior. Its performance is evaluated on the MVSEC dataset, where it achieves 7.9e-3 and 5.9e-3 RMS relative error on two independent sequences, outperforming the existing publicly available event-based stereo VO pipeline by two and four times, respectively.
comment: Submitted to IEEE Robotics and Automation Letters (RA-L). Manuscript #23-1314. 8 pages, 4 figures
☆ Cooperative Hardware-Prompt Learning for Snapshot Compressive Imaging
Snapshot compressive imaging emerges as a promising technology for acquiring real-world hyperspectral signals. It uses an optical encoder and compressively produces the 2D measurement, followed by which the 3D hyperspectral data can be retrieved via training a deep reconstruction network. Existing reconstruction models are trained with a single hardware instance, whose performance is vulnerable to hardware perturbation or replacement, demonstrating an overfitting issue to the physical configuration. This defect limits the deployment of pre-trained models since they would suffer from large performance degradation when are assembled to unseen hardware. To better facilitate the reconstruction model with new hardware, previous efforts resort to centralized training by collecting multi-hardware and data, which is impractical when dealing with proprietary assets among institutions. In light of this, federated learning (FL) has become a feasible solution to enable cross-hardware cooperation without breaking privacy. However, the naive FedAvg is subject to client drift upon data heterogeneity owning to the hardware inconsistency. In this work, we tackle this challenge by marrying prompt tuning with FL to snapshot compressive imaging for the first time and propose an federated hardware-prompt learning (FedHP) method. Rather than mitigating the client drift by rectifying the gradients, which only takes effect on the learning manifold but fails to touch the heterogeneity rooted in the input data space, the proposed FedHP globally learns a hardware-conditioned prompter to align the data distribution, which serves as an indicator of the data inconsistency stemming from different pre-defined coded apertures. Extensive experiments demonstrate that the proposed method well coordinates the pre-trained model to indeterminate hardware configurations.
comment: 11 figures, 4 tables
☆ Addressing Discrepancies in Semantic and Visual Alignment in Neural Networks
For the task of image classification, neural networks primarily rely on visual patterns. In robust networks, we would expect for visually similar classes to be represented similarly. We consider the problem of when semantically similar classes are visually dissimilar, and when visual similarity is present among non-similar classes. We propose a data augmentation technique with the goal of better aligning semantically similar classes with arbitrary (non-visual) semantic relationships. We leverage recent work in diffusion-based semantic mixing to generate semantic hybrids of two classes, and these hybrids are added to the training set as augmented data. We evaluate whether the method increases semantic alignment by evaluating model performance on adversarially perturbed data, with the idea that it should be easier for an adversary to switch one class to a similarly represented class. Results demonstrate that there is an increase in alignment of semantically similar classes when using our proposed data augmentation method.
☆ Privacy-Preserving Remote Heart Rate Estimation from Facial Videos
Remote Photoplethysmography (rPPG) is the process of estimating PPG from facial videos. While this approach benefits from contactless interaction, it is reliant on videos of faces, which often constitutes an important privacy concern. Recent research has revealed that deep learning techniques are vulnerable to attacks, which can result in significant data breaches making deep rPPG estimation even more sensitive. To address this issue, we propose a data perturbation method that involves extraction of certain areas of the face with less identity-related information, followed by pixel shuffling and blurring. Our experiments on two rPPG datasets (PURE and UBFC) show that our approach reduces the accuracy of facial recognition algorithms by over 60%, with minimal impact on rPPG extraction. We also test our method on three facial recognition datasets (LFW, CALFW, and AgeDB), where our approach reduced performance by nearly 50%. Our findings demonstrate the potential of our approach as an effective privacy-preserving solution for rPPG estimation.
comment: Accepted in IEEE International Conference on Systems, Man, and Cybernetics (SMC) 2023
☆ Reconstruction Distortion of Learned Image Compression with Imperceptible Perturbations
Learned Image Compression (LIC) has recently become the trending technique for image transmission due to its notable performance. Despite its popularity, the robustness of LIC with respect to the quality of image reconstruction remains under-explored. In this paper, we introduce an imperceptible attack approach designed to effectively degrade the reconstruction quality of LIC, resulting in the reconstructed image being severely disrupted by noise where any object in the reconstructed images is virtually impossible. More specifically, we generate adversarial examples by introducing a Frobenius norm-based loss function to maximize the discrepancy between original images and reconstructed adversarial examples. Further, leveraging the insensitivity of high-frequency components to human vision, we introduce Imperceptibility Constraint (IC) to ensure that the perturbations remain inconspicuous. Experiments conducted on the Kodak dataset using various LIC models demonstrate effectiveness. In addition, we provide several findings and suggestions for designing future defenses.
comment: 7 pages
☆ Exploring the Versatility of Zero-Shot CLIP for Interstitial Lung Disease Classification
Interstitial lung diseases (ILD) present diagnostic challenges due to their varied manifestations and overlapping imaging features. To address this, we propose a machine learning approach that utilizes CLIP, a multimodal (image and text) self-supervised model, for ILD classification. We extensively integrate zero-shot CLIP throughout our workflow, starting from the initial extraction of image patches from volumetric CT scans and proceeding to ILD classification using "patch montages". Furthermore, we investigate how domain adaptive pretraining (DAPT) CLIP with task-specific images (CT "patch montages" extracted with ILD-specific prompts for CLIP) and/or text (lung-specific sections of radiology reports) affects downstream ILD classification performance. By leveraging CLIP-extracted "patch montages" and DAPT, we achieve strong zero-shot ILD classification results, including an AUROC of 0.893, without the need for any labeled training data. This work highlights the versatility and potential of multimodal models like CLIP for medical image classification tasks where labeled data is scarce.
comment: 11 pages, 11 figures
☆ 4DSR-GCN: 4D Video Point Cloud Upsampling using Graph Convolutional Networks
Time varying sequences of 3D point clouds, or 4D point clouds, are now being acquired at an increasing pace in several applications (e.g., LiDAR in autonomous or assisted driving). In many cases, such volume of data is transmitted, thus requiring that proper compression tools are applied to either reduce the resolution or the bandwidth. In this paper, we propose a new solution for upscaling and restoration of time-varying 3D video point clouds after they have been heavily compressed. In consideration of recent growing relevance of 3D applications, %We focused on a model allowing user-side upscaling and artifact removal for 3D video point clouds, a real-time stream of which would require . Our model consists of a specifically designed Graph Convolutional Network (GCN) that combines Dynamic Edge Convolution and Graph Attention Networks for feature aggregation in a Generative Adversarial setting. By taking inspiration PointNet++, We present a different way to sample dense point clouds with the intent to make these modules work in synergy to provide each node enough features about its neighbourhood in order to later on generate new vertices. Compared to other solutions in the literature that address the same task, our proposed model is capable of obtaining comparable results in terms of quality of the reconstruction, while using a substantially lower number of parameters (about 300KB), making our solution deployable in edge computing devices such as LiDAR.
☆ Pedestrian Crossing Action Recognition and Trajectory Prediction with 3D Human Keypoints ICRA 2023
Accurate understanding and prediction of human behaviors are critical prerequisites for autonomous vehicles, especially in highly dynamic and interactive scenarios such as intersections in dense urban areas. In this work, we aim at identifying crossing pedestrians and predicting their future trajectories. To achieve these goals, we not only need the context information of road geometry and other traffic participants but also need fine-grained information of the human pose, motion and activity, which can be inferred from human keypoints. In this paper, we propose a novel multi-task learning framework for pedestrian crossing action recognition and trajectory prediction, which utilizes 3D human keypoints extracted from raw sensor data to capture rich information on human pose and activity. Moreover, we propose to apply two auxiliary tasks and contrastive learning to enable auxiliary supervisions to improve the learned keypoints representation, which further enhances the performance of major tasks. We validate our approach on a large-scale in-house dataset, as well as a public benchmark dataset, and show that our approach achieves state-of-the-art performance on a wide range of evaluation metrics. The effectiveness of each model component is validated in a detailed ablation study.
comment: ICRA 2023
☆ Pseudo Labels for Single Positive Multi-Label Learning ICLR 2023
The cost of data annotation is a substantial impediment for multi-label image classification: in every image, every category must be labeled as present or absent. Single positive multi-label (SPML) learning is a cost-effective solution, where models are trained on a single positive label per image. Thus, SPML is a more challenging domain, since it requires dealing with missing labels. In this work, we propose a method to turn single positive data into fully-labeled data: Pseudo Multi-Labels. Basically, a teacher network is trained on single positive labels. Then, we treat the teacher model's predictions on the training data as ground-truth labels to train a student network on fully-labeled images. With this simple approach, we show that the performance achieved by the student model approaches that of a model trained on the actual fully-labeled images.
comment: ICLR 2023, Tiny Papers Track
♻ ☆ D$^2$CSG: Unsupervised Learning of Compact CSG Trees with Dual Complements and Dropouts
We present D$^2$CSG, a neural model composed of two dual and complementary network branches, with dropouts, for unsupervised learning of compact constructive solid geometry (CSG) representations of 3D CAD shapes. Our network is trained to reconstruct a 3D shape by a fixed-order assembly of quadric primitives, with both branches producing a union of primitive intersections or inverses. A key difference between D$^2$CSG and all prior neural CSG models is its dedicated residual branch to assemble the potentially complex shape complement, which is subtracted from an overall shape modeled by the cover branch. With the shape complements, our network is provably general, while the weight dropout further improves compactness of the CSG tree by removing redundant primitives. We demonstrate both quantitatively and qualitatively that D$^2$CSG produces compact CSG reconstructions with superior quality and more natural primitives than all existing alternatives, especially over complex and high-genus CAD shapes.
comment: 9 pages
♻ ☆ Instruct-NeRF2NeRF: Editing 3D Scenes with Instructions
We propose a method for editing NeRF scenes with text-instructions. Given a NeRF of a scene and the collection of images used to reconstruct it, our method uses an image-conditioned diffusion model (InstructPix2Pix) to iteratively edit the input images while optimizing the underlying scene, resulting in an optimized 3D scene that respects the edit instruction. We demonstrate that our proposed method is able to edit large-scale, real-world scenes, and is able to accomplish more realistic, targeted edits than prior work.
comment: Project website: https://instruct-nerf2nerf.github.io; v1. Revisions to related work and discussion
♻ ☆ VOCALExplore: Pay-as-You-Go Video Data Exploration and Model Building [Technical Report]
We introduce VOCALExplore, a system designed to support users in building domain-specific models over video datasets. VOCALExplore supports interactive labeling sessions and trains models using user-supplied labels. VOCALExplore maximizes model quality by automatically deciding how to select samples based on observed skew in the collected labels. It also selects the optimal video representations to use when training models by casting feature selection as a rising bandit problem. Finally, VOCALExplore implements optimizations to achieve low latency without sacrificing model performance. We demonstrate that VOCALExplore achieves close to the best possible model quality given candidate acquisition functions and feature extractors, and it does so with low visible latency (~1 second per iteration) and no expensive preprocessing.
♻ ☆ Continual Vision-Language Representation Learning with Off-Diagonal Information
Large-scale multi-modal contrastive learning frameworks like CLIP typically require a large amount of image-text samples for training. However, these samples are always collected continuously in real scenarios. This paper discusses the feasibility of continual CLIP training using streaming data. Unlike continual learning based on self-supervised learning methods for pure images, which is empirically robust against catastrophic forgetting, CLIP's performance degeneration in the continual setting is significant and non-neglectable. By analyzing the changes in the model's representation space during continual CLIP training from a spatial geometry perspective, we explore and summarize these spatial variations as Spatial Disorder (SD), which can be divided into Intra-modal Rotation and Inter-modal Deviation. Moreover, we empirically and theoretically demonstrate how SD leads to a performance decline for CLIP on cross-modal retrieval tasks. To alleviate SD, we propose a new continual vision-language representation learning framework Mod-X: Maintain off-diagonal information-matriX. By selectively aligning the off-diagonal information distribution of contrastive matrices, the Mod-X improves the capability of the multi-modal model by maintaining the multi-modal representation space alignment on the old data domain during continuously fitting the new training data domain. Experiments on commonly used datasets with different scales and scopes have demonstrated the effectiveness of our method.
♻ ☆ Dense and Aligned Captions (DAC) Promote Compositional Reasoning in VL Models
Vision and Language (VL) models offer an effective method for aligning representation spaces of images and text, leading to numerous applications such as cross-modal retrieval, visual question answering, captioning, and more. However, the aligned image-text spaces learned by all the popular VL models are still suffering from the so-called `object bias' - their representations behave as `bags of nouns', mostly ignoring or downsizing the attributes, relations, and states of objects described/appearing in texts/images. Although some great attempts at fixing these `compositional reasoning' issues were proposed in the recent literature, the problem is still far from being solved. In this paper, we uncover two factors limiting the VL models' compositional reasoning performance. These two factors are properties of the paired VL dataset used for finetuning and pre-training the VL model: (i) the caption quality, or in other words `image-alignment', of the texts; and (ii) the `density' of the captions in the sense of mentioning all the details appearing on the image. We propose a fine-tuning approach for automatically treating these factors leveraging a standard VL dataset (CC3M). Applied to CLIP, we demonstrate its significant compositional reasoning performance increase of up to $\sim27\%$ over the base model, up to $\sim20\%$ over the strongest baseline, and by $6.7\%$ on average.
♻ ☆ Grounding Language Models to Images for Multimodal Inputs and Outputs ICML 2023
We propose an efficient method to ground pretrained text-only language models to the visual domain, enabling them to process arbitrarily interleaved image-and-text data, and generate text interleaved with retrieved images. Our method leverages the abilities of language models learnt from large scale text-only pretraining, such as in-context learning and free-form text generation. We keep the language model frozen, and finetune input and output linear layers to enable cross-modality interactions. This allows our model to process arbitrarily interleaved image-and-text inputs, and generate free-form text interleaved with retrieved images. We achieve strong zero-shot performance on grounded tasks such as contextual image retrieval and multimodal dialogue, and showcase compelling interactive abilities. Our approach works with any off-the-shelf language model and paves the way towards an effective, general solution for leveraging pretrained language models in visually grounded settings.
comment: Published in ICML 2023. Project page: https://jykoh.com/fromage
♻ ☆ CommonScenes: Generating Commonsense 3D Indoor Scenes with Scene Graphs
Controllable scene synthesis aims to create interactive environments for various industrial use cases. Scene graphs provide a highly suitable interface to facilitate these applications by abstracting the scene context in a compact manner. Existing methods, reliant on retrieval from extensive databases or pre-trained shape embeddings, often overlook scene-object and object-object relationships, leading to inconsistent results due to their limited generation capacity. To address this issue, we present CommonScenes, a fully generative model that converts scene graphs into corresponding controllable 3D scenes, which are semantically realistic and conform to commonsense. Our pipeline consists of two branches, one predicting the overall scene layout via a variational auto-encoder and the other generating compatible shapes via latent diffusion, capturing global scene-object and local inter-object relationships while preserving shape diversity. The generated scenes can be manipulated by editing the input scene graph and sampling the noise in the diffusion model. Due to lacking a scene graph dataset offering high-quality object-level meshes with relations, we also construct SG-FRONT, enriching the off-the-shelf indoor dataset 3D-FRONT with additional scene graph labels. Extensive experiments are conducted on SG-FRONT where CommonScenes shows clear advantages over other methods regarding generation consistency, quality, and diversity. Codes and the dataset will be released upon acceptance.
♻ ☆ A Closer Look at Few-shot Classification Again ICML 2023
Few-shot classification consists of a training phase where a model is learned on a relatively large dataset and an adaptation phase where the learned model is adapted to previously-unseen tasks with limited labeled samples. In this paper, we empirically prove that the training algorithm and the adaptation algorithm can be completely disentangled, which allows algorithm analysis and design to be done individually for each phase. Our meta-analysis for each phase reveals several interesting insights that may help better understand key aspects of few-shot classification and connections with other fields such as visual representation learning and transfer learning. We hope the insights and research challenges revealed in this paper can inspire future work in related directions. Code and pre-trained models (in PyTorch) are available at https://github.com/Frankluox/CloserLookAgainFewShot.
comment: Accepted at ICML 2023
♻ ☆ Learning Runtime Decisions for Adaptive Real-Time Perception
Real-time perception requires planned resource utilization. Computational planning in real-time perception is governed by two considerations -- accuracy and latency. There exist run-time decisions (e.g. choice of input resolution) that induce tradeoffs affecting performance on a given hardware, arising from intrinsic (content, e.g. scene clutter) and extrinsic (system, e.g. resource contention) characteristics. Earlier runtime execution frameworks employed rule-based decision algorithms and operated with a fixed algorithm latency budget to balance these concerns, which is sub-optimal and inflexible. We propose Chanakya, a learned approximate execution framework that naturally derives from the streaming perception paradigm, to automatically learn decisions induced by these tradeoffs instead. Chanakya is trained via novel rewards balancing accuracy and latency implicitly, without approximating either objectives. Chanakya simultaneously considers intrinsic and extrinsic context, and predicts decisions in a flexible manner. Chanakya, designed with low overhead in mind, outperforms state-of-the-art static and dynamic execution policies on public datasets on both server GPUs and edge devices.
comment: 18 pages
♻ ☆ Best Prompts for Text-to-Image Models and How to Find Them SIGIR '23
Recent progress in generative models, especially in text-guided diffusion models, has enabled the production of aesthetically-pleasing imagery resembling the works of professional human artists. However, one has to carefully compose the textual description, called the prompt, and augment it with a set of clarifying keywords. Since aesthetics are challenging to evaluate computationally, human feedback is needed to determine the optimal prompt formulation and keyword combination. In this paper, we present a human-in-the-loop approach to learning the most useful combination of prompt keywords using a genetic algorithm. We also show how such an approach can improve the aesthetic appeal of images depicting the same descriptions.
comment: 13 pages (6 main pages), 7 figures, 4 tables, accepted at SIGIR '23 Short Paper Track
♻ ☆ Automatic Photo Orientation Detection with Convolutional Neural Networks
We apply convolutional neural networks (CNN) to the problem of image orientation detection in the context of determining the correct orientation (from 0, 90, 180, and 270 degrees) of a consumer photo. The problem is especially important for digitazing analog photographs. We substantially improve on the published state of the art in terms of the performance on one of the standard datasets, and test our system on a more difficult large dataset of consumer photos. We use Guided Backpropagation to obtain insights into how our CNN detects photo orientation, and to explain its mistakes.
♻ ☆ Dual Relation Knowledge Distillation for Object Detection IJCAI-2023
Knowledge distillation is an effective method for model compression. However, it is still a challenging topic to apply knowledge distillation to detection tasks. There are two key points resulting in poor distillation performance for detection tasks. One is the serious imbalance between foreground and background features, another one is that small object lacks enough feature representation. To solve the above issues, we propose a new distillation method named dual relation knowledge distillation (DRKD), including pixel-wise relation distillation and instance-wise relation distillation. The pixel-wise relation distillation embeds pixel-wise features in the graph space and applies graph convolution to capture the global pixel relation. By distilling the global pixel relation, the student detector can learn the relation between foreground and background features, and avoid the difficulty of distilling features directly for the feature imbalance issue. Besides, we find that instance-wise relation supplements valuable knowledge beyond independent features for small objects. Thus, the instance-wise relation distillation is designed, which calculates the similarity of different instances to obtain a relation matrix. More importantly, a relation filter module is designed to highlight valuable instance relations. The proposed dual relation knowledge distillation is general and can be easily applied for both one-stage and two-stage detectors. Our method achieves state-of-the-art performance, which improves Faster R-CNN based on ResNet50 from 38.4% to 41.6% mAP and improves RetinaNet based on ResNet50 from 37.4% to 40.3% mAP on COCO 2017.
comment: Accepted by IJCAI-2023
♻ ☆ [Work in progress] Scalable, out-of-the box segmentation of individual particles from mineral samples acquired with micro CT
Minerals are indispensable for a functioning modern society. Yet, their supply is limited causing a need for optimizing their exploration and extraction both from ores and recyclable materials. Typically, these processes must be meticulously adapted to the precise properties of the processed particles, an extensive characterization of their shapes, appearances as well as the overall material composition. Current approaches perform this analysis based on bulk segmentation and characterization of particles imaged with a micro CT, and rely on rudimentary postprocessing techniques to separate touching particles. However, due to their inability to reliably perform this separation as well as the need to retrain or reconfigure methods for each new image, these approaches leave untapped potential to be leveraged. Here, we propose ParticleSeg3D, an instance segmentation method that is able to extract individual particles from large micro CT images taken from mineral samples embedded in an epoxy matrix. Our approach is based on the powerful nnU-Net framework, introduces a particle size normalization, makes use of a border-core representation to enable instance segmentation and is trained with a large dataset containing particles of numerous different materials and minerals. We demonstrate that ParticleSeg3D can be applied out-of-the box to a large variety of particle types, including materials and appearances that have not been part of the training set. Thus, no further manual annotations and retraining are required when applying the method to new mineral samples, enabling substantially higher scalability of experiments than existing methods. Our code and dataset are made publicly available.
♻ ☆ AirfRANS: High Fidelity Computational Fluid Dynamics Dataset for Approximating Reynolds-Averaged Navier-Stokes Solutions
Surrogate models are necessary to optimize meaningful quantities in physical dynamics as their recursive numerical resolutions are often prohibitively expensive. It is mainly the case for fluid dynamics and the resolution of Navier-Stokes equations. However, despite the fast-growing field of data-driven models for physical systems, reference datasets representing real-world phenomena are lacking. In this work, we develop AirfRANS, a dataset for studying the two-dimensional incompressible steady-state Reynolds-Averaged Navier-Stokes equations over airfoils at a subsonic regime and for different angles of attacks. We also introduce metrics on the stress forces at the surface of geometries and visualization of boundary layers to assess the capabilities of models to accurately predict the meaningful information of the problem. Finally, we propose deep learning baselines on four machine learning tasks to study AirfRANS under different constraints for generalization considerations: big and scarce data regime, Reynolds number, and angle of attack extrapolation.
♻ ☆ Provably Learning Diverse Features in Multi-View Data with Midpoint Mixup ICML 2023
Mixup is a data augmentation technique that relies on training using random convex combinations of data points and their labels. In recent years, Mixup has become a standard primitive used in the training of state-of-the-art image classification models due to its demonstrated benefits over empirical risk minimization with regards to generalization and robustness. In this work, we try to explain some of this success from a feature learning perspective. We focus our attention on classification problems in which each class may have multiple associated features (or views) that can be used to predict the class correctly. Our main theoretical results demonstrate that, for a non-trivial class of data distributions with two features per class, training a 2-layer convolutional network using empirical risk minimization can lead to learning only one feature for almost all classes while training with a specific instantiation of Mixup succeeds in learning both features for every class. We also show empirically that these theoretical insights extend to the practical settings of image benchmarks modified to have multiple features.
comment: 37 pages, 2 figures, ICML 2023
♻ ☆ GPA-Net:No-Reference Point Cloud Quality Assessment with Multi-task Graph Convolutional Network
With the rapid development of 3D vision, point cloud has become an increasingly popular 3D visual media content. Due to the irregular structure, point cloud has posed novel challenges to the related research, such as compression, transmission, rendering and quality assessment. In these latest researches, point cloud quality assessment (PCQA) has attracted wide attention due to its significant role in guiding practical applications, especially in many cases where the reference point cloud is unavailable. However, current no-reference metrics which based on prevalent deep neural network have apparent disadvantages. For example, to adapt to the irregular structure of point cloud, they require preprocessing such as voxelization and projection that introduce extra distortions, and the applied grid-kernel networks, such as Convolutional Neural Networks, fail to extract effective distortion-related features. Besides, they rarely consider the various distortion patterns and the philosophy that PCQA should exhibit shifting, scaling, and rotational invariance. In this paper, we propose a novel no-reference PCQA metric named the Graph convolutional PCQA network (GPA-Net). To extract effective features for PCQA, we propose a new graph convolution kernel, i.e., GPAConv, which attentively captures the perturbation of structure and texture. Then, we propose the multi-task framework consisting of one main task (quality regression) and two auxiliary tasks (distortion type and degree predictions). Finally, we propose a coordinate normalization module to stabilize the results of GPAConv under shift, scale and rotation transformations. Experimental results on two independent databases show that GPA-Net achieves the best performance compared to the state-of-the-art no-reference PCQA metrics, even better than some full-reference metrics in some cases.
♻ ☆ GaitGS: Temporal Feature Learning in Granularity and Span Dimension for Gait Recognition
Gait recognition is an emerging biological recognition technology that identifies and verifies individuals based on their walking patterns. However, many current methods are limited in their use of temporal information. In order to fully harness the potential of gait recognition, it is crucial to consider temporal features at various granularities and spans. Hence, in this paper, we propose a novel framework named GaitGS, which aggregates temporal features in the granularity dimension and span dimension simultaneously. Specifically, Multi-Granularity Feature Extractor (MGFE) is proposed to focus on capturing the micro-motion and macro-motion information at the frame level and unit level respectively. Moreover, we present Multi-Span Feature Learning (MSFL) module to generate global and local temporal representations. On three popular gait datasets, extensive experiments demonstrate the state-of-the-art performance of our method. Our method achieves the Rank-1 accuracies of 92.9% (+0.5%), 52.0% (+1.4%), and 97.5% (+0.8%) on CASIA-B, GREW, and OU-MVLP respectively. The source code will be released soon.
comment: 14 pages, 6 figures
♻ ☆ Segment Anything in 3D with NeRFs
Recently, the Segment Anything Model (SAM) emerged as a powerful vision foundation model which is capable to segment anything in 2D images. This paper aims to generalize SAM to segment 3D objects. Rather than replicating the data acquisition and annotation procedure which is costly in 3D, we design an efficient solution, leveraging the Neural Radiance Field (NeRF) as a cheap and off-the-shelf prior that connects multi-view 2D images to the 3D space. We refer to the proposed solution as SA3D, for Segment Anything in 3D. It is only required to provide a manual segmentation prompt (e.g., rough points) for the target object in a single view, which is used to generate its 2D mask in this view with SAM. Next, SA3D alternately performs mask inverse rendering and cross-view self-prompting across various views to iteratively complete the 3D mask of the target object constructed with voxel grids. The former projects the 2D mask obtained by SAM in the current view onto 3D mask with guidance of the density distribution learned by the NeRF; The latter extracts reliable prompts automatically as the input to SAM from the NeRF-rendered 2D mask in another view. We show in experiments that SA3D adapts to various scenes and achieves 3D segmentation within minutes. Our research offers a generic and efficient methodology to lift a 2D vision foundation model to 3D, as long as the 2D model can steadily address promptable segmentation across multiple views. The project page is at https://jumpat.github.io/SA3D/.
comment: Work in progress. Project page: https://jumpat.github.io/SA3D/
♻ ☆ Curricular Contrastive Regularization for Physics-aware Single Image Dehazing CVPR2023
Considering the ill-posed nature, contrastive regularization has been developed for single image dehazing, introducing the information from negative images as a lower bound. However, the contrastive samples are nonconsensual, as the negatives are usually represented distantly from the clear (i.e., positive) image, leaving the solution space still under-constricted. Moreover, the interpretability of deep dehazing models is underexplored towards the physics of the hazing process. In this paper, we propose a novel curricular contrastive regularization targeted at a consensual contrastive space as opposed to a non-consensual one. Our negatives, which provide better lower-bound constraints, can be assembled from 1) the hazy image, and 2) corresponding restorations by other existing methods. Further, due to the different similarities between the embeddings of the clear image and negatives, the learning difficulty of the multiple components is intrinsically imbalanced. To tackle this issue, we customize a curriculum learning strategy to reweight the importance of different negatives. In addition, to improve the interpretability in the feature space, we build a physics-aware dual-branch unit according to the atmospheric scattering model. With the unit, as well as curricular contrastive regularization, we establish our dehazing network, named C2PNet. Extensive experiments demonstrate that our C2PNet significantly outperforms state-of-the-art methods, with extreme PSNR boosts of 3.94dB and 1.50dB, respectively, on SOTS-indoor and SOTS-outdoor datasets.
comment: This paper is accepted by CVPR2023
♻ ☆ Hybrid Energy Based Model in the Feature Space for Out-of-Distribution Detection
Out-of-distribution (OOD) detection is a critical requirement for the deployment of deep neural networks. This paper introduces the HEAT model, a new post-hoc OOD detection method estimating the density of in-distribution (ID) samples using hybrid energy-based models (EBM) in the feature space of a pre-trained backbone. HEAT complements prior density estimators of the ID density, e.g. parametric models like the Gaussian Mixture Model (GMM), to provide an accurate yet robust density estimation. A second contribution is to leverage the EBM framework to provide a unified density estimation and to compose several energy terms. Extensive experiments demonstrate the significance of the two contributions. HEAT sets new state-of-the-art OOD detection results on the CIFAR-10 / CIFAR-100 benchmark as well as on the large-scale Imagenet benchmark. The code is available at: https://github.com/MarcLafon/heatood.
♻ ☆ Nonlinear Advantage: Trained Networks Might Not Be As Complex as You Think
We perform an empirical study of the behaviour of deep networks when fully linearizing some of its feature channels through a sparsity prior on the overall number of nonlinear units in the network. In experiments on image classification and machine translation tasks, we investigate how much we can simplify the network function towards linearity before performance collapses. First, we observe a significant performance gap when reducing nonlinearity in the network function early on as opposed to late in training, in-line with recent observations on the time-evolution of the data-dependent NTK. Second, we find that after training, we are able to linearize a significant number of nonlinear units while maintaining a high performance, indicating that much of a network's expressivity remains unused but helps gradient descent in early stages of training. To characterize the depth of the resulting partially linearized network, we introduce a measure called average path length, representing the average number of active nonlinearities encountered along a path in the network graph. Under sparsity pressure, we find that the remaining nonlinear units organize into distinct structures, forming core-networks of near constant effective depth and width, which in turn depend on task difficulty.
♻ ☆ Mechanistic Mode Connectivity ICML
We study neural network loss landscapes through the lens of mode connectivity, the observation that minimizers of neural networks retrieved via training on a dataset are connected via simple paths of low loss. Specifically, we ask the following question: are minimizers that rely on different mechanisms for making their predictions connected via simple paths of low loss? We provide a definition of mechanistic similarity as shared invariances to input transformations and demonstrate that lack of linear connectivity between two models implies they use dissimilar mechanisms for making their predictions. Relevant to practice, this result helps us demonstrate that naive fine-tuning on a downstream dataset can fail to alter a model's mechanisms, e.g., fine-tuning can fail to eliminate a model's reliance on spurious attributes. Our analysis also motivates a method for targeted alteration of a model's mechanisms, named connectivity-based fine-tuning (CBFT), which we analyze using several synthetic datasets for the task of reducing a model's reliance on spurious attributes.
comment: ICML, 2023
♻ ☆ Joint Adaptive Representations for Image-Language Learning
Image-language learning has made unprecedented progress in visual understanding. These developments have come at high costs, as contemporary vision-language models require large model scales and amounts of data. We here propose a much easier recipe for image-language learning, which produces effective models, outperforming bigger and more expensive ones, often trained on orders of magnitude larger datasets. Our key finding is the joint learning of a compact vision and language representation, which adaptively and iteratively fuses the multi-modal features. This results in a more effective image-language learning, greatly lowering the FLOPs by combining and reducing the number of tokens for both text and images, e.g. a 33\% reduction in FLOPs is achieved, compared to baseline fusion techniques used by popular image-language models, while improving performance. This also allows the model to scale without a large increase in FLOPs or memory. In addition, we propose adaptive pre-training data sampling which improves the data efficiency. The proposed approach achieves competitive performance compared to much larger models, and does so with significantly less data and FLOPs. With only 40M training examples and with 39 GFLOPs our lightweight model outperforms many times larger state-of-the-art models of 2-20x more FLOPs and using bigger datasets some of which with close to 1B training examples.
comment: T4V Workshop
♻ ☆ Toward Foundation Models for Earth Monitoring: Generalizable Deep Learning Models for Natural Hazard Segmentation
Climate change results in an increased probability of extreme weather events that put societies and businesses at risk on a global scale. Therefore, near real-time mapping of natural hazards is an emerging priority for the support of natural disaster relief, risk management, and informing governmental policy decisions. Recent methods to achieve near real-time mapping increasingly leverage deep learning (DL). However, DL-based approaches are designed for one specific task in a single geographic region based on specific frequency bands of satellite data. Therefore, DL models used to map specific natural hazards struggle with their generalization to other types of natural hazards in unseen regions. In this work, we propose a methodology to significantly improve the generalizability of DL natural hazards mappers based on pre-training on a suitable pre-task. Without access to any data from the target domain, we demonstrate this improved generalizability across four U-Net architectures for the segmentation of unseen natural hazards. Importantly, our method is invariant to geographic differences and differences in the type of frequency bands of satellite data. By leveraging characteristics of unlabeled images from the target domain that are publicly available, our approach is able to further improve the generalization behavior without fine-tuning. Thereby, our approach supports the development of foundation models for earth monitoring with the objective of directly segmenting unseen natural hazards across novel geographic regions given different sources of satellite imagery.
comment: Accepted at IEEE International Geoscience and Remote Sensing Symposium (IGARSS 2023)
♻ ☆ Tree-Ring Watermarks: Fingerprints for Diffusion Images that are Invisible and Robust
Watermarking the outputs of generative models is a crucial technique for tracing copyright and preventing potential harm from AI-generated content. In this paper, we introduce a novel technique called Tree-Ring Watermarking that robustly fingerprints diffusion model outputs. Unlike existing methods that perform post-hoc modifications to images after sampling, Tree-Ring Watermarking subtly influences the entire sampling process, resulting in a model fingerprint that is invisible to humans. The watermark embeds a pattern into the initial noise vector used for sampling. These patterns are structured in Fourier space so that they are invariant to convolutions, crops, dilations, flips, and rotations. After image generation, the watermark signal is detected by inverting the diffusion process to retrieve the noise vector, which is then checked for the embedded signal. We demonstrate that this technique can be easily applied to arbitrary diffusion models, including text-conditioned Stable Diffusion, as a plug-in with negligible loss in FID. Our watermark is semantically hidden in the image space and is far more robust than watermarking alternatives that are currently deployed. Code is available at https://github.com/YuxinWenRick/tree-ring-watermark.
comment: 16 pages, 8 figures, code is available at https://github.com/YuxinWenRick/tree-ring-watermark, fixed the repo link
♻ ☆ Better Diffusion Models Further Improve Adversarial Training ICML 2023
It has been recognized that the data generated by the denoising diffusion probabilistic model (DDPM) improves adversarial training. After two years of rapid development in diffusion models, a question naturally arises: can better diffusion models further improve adversarial training? This paper gives an affirmative answer by employing the most recent diffusion model which has higher efficiency ($\sim 20$ sampling steps) and image quality (lower FID score) compared with DDPM. Our adversarially trained models achieve state-of-the-art performance on RobustBench using only generated data (no external datasets). Under the $\ell_\infty$-norm threat model with $\epsilon=8/255$, our models achieve $70.69\%$ and $42.67\%$ robust accuracy on CIFAR-10 and CIFAR-100, respectively, i.e. improving upon previous state-of-the-art models by $+4.58\%$ and $+8.03\%$. Under the $\ell_2$-norm threat model with $\epsilon=128/255$, our models achieve $84.86\%$ on CIFAR-10 ($+4.44\%$). These results also beat previous works that use external data. We also provide compelling results on the SVHN and TinyImageNet datasets. Our code is available at https://github.com/wzekai99/DM-Improves-AT.
comment: ICML 2023
♻ ☆ Boosting Text-to-Image Diffusion Models with Fine-Grained Semantic Rewards
Recent advances in text-to-image diffusion models have achieved remarkable success in generating high-quality, realistic images from given text prompts. However, previous methods fail to perform accurate modality alignment between text concepts and generated images due to the lack of fine-level semantic guidance that successfully diagnoses the modality discrepancy. In this paper, we propose FineRewards to improve the alignment between text and images in text-to-image diffusion models by introducing two new fine-grained semantic rewards: the caption reward and the Semantic Segment Anything (SAM) reward. From the global semantic view, the caption reward generates a corresponding detailed caption that depicts all important contents in the synthetic image via a BLIP-2 model and then calculates the reward score by measuring the similarity between the generated caption and the given prompt. From the local semantic view, the SAM reward segments the generated images into local parts with category labels, and scores the segmented parts by measuring the likelihood of each category appearing in the prompted scene via a large language model, i.e., Vicuna-7B. Additionally, we adopt an assemble reward-ranked learning strategy to enable the integration of multiple reward functions to jointly guide the model training. Adapting results of text-to-image models on the MS-COCO benchmark show that the proposed semantic reward outperforms other baseline reward functions with a considerable margin on both visual quality and semantic similarity with the input prompt. Moreover, by adopting the assemble reward-ranked learning strategy, we further demonstrate that model performance is further improved when adapting under the unifying of the proposed semantic reward with the current image rewards.
♻ ☆ Realistic Full-Body Anonymization with Surface-Guided GANs WACV 2023
Recent work on image anonymization has shown that generative adversarial networks (GANs) can generate near-photorealistic faces to anonymize individuals. However, scaling up these networks to the entire human body has remained a challenging and yet unsolved task. We propose a new anonymization method that generates realistic humans for in-the-wild images. A key part of our design is to guide adversarial nets by dense pixel-to-surface correspondences between an image and a canonical 3D surface. We introduce Variational Surface-Adaptive Modulation (V-SAM) that embeds surface information throughout the generator. Combining this with our novel discriminator surface supervision loss, the generator can synthesize high quality humans with diverse appearances in complex and varying scenes. We demonstrate that surface guidance significantly improves image quality and diversity of samples, yielding a highly practical generator. Finally, we show that our method preserves data usability without infringing privacy when collecting image datasets for training computer vision models. Source code and appendix is available at: \href{https://github.com/hukkelas/full_body_anonymization}{github.com/hukkelas/full\_body\_anonymization}
comment: 8 pages, 7 figures, 6 tables. Source code and appendix available at: https://www.github.com/hukkelas/full_body_anonymization. Published at WACV 2023
♻ ☆ Ghost in the Minecraft: Generally Capable Agents for Open-World Environments via Large Language Models with Text-based Knowledge and Memory
The captivating realm of Minecraft has attracted substantial research interest in recent years, serving as a rich platform for developing intelligent agents capable of functioning in open-world environments. However, the current research landscape predominantly focuses on specific objectives, such as the popular "ObtainDiamond" task, and has not yet shown effective generalization to a broader spectrum of tasks. Furthermore, the current leading success rate for the "ObtainDiamond" task stands at around 20%, highlighting the limitations of Reinforcement Learning (RL) based controllers used in existing methods. To tackle these challenges, we introduce Ghost in the Minecraft (GITM), a novel framework integrates Large Language Models (LLMs) with text-based knowledge and memory, aiming to create Generally Capable Agents (GCAs) in Minecraft. These agents, equipped with the logic and common sense capabilities of LLMs, can skillfully navigate complex, sparse-reward environments with text-based interactions. We develop a set of structured actions and leverage LLMs to generate action plans for the agents to execute. The resulting LLM-based agent markedly surpasses previous methods, achieving a remarkable improvement of +47.5% in success rate on the "ObtainDiamond" task, demonstrating superior robustness compared to traditional RL-based controllers. Notably, our agent is the first to procure all items in the Minecraft Overworld technology tree, demonstrating its extensive capabilities. GITM does not need any GPU for training, but a single CPU node with 32 CPU cores is enough. This research shows the potential of LLMs in developing capable agents for handling long-horizon, complex tasks and adapting to uncertainties in open-world environments. See the project website at https://github.com/OpenGVLab/GITM.
♻ ☆ MCDIP-ADMM: Overcoming Overfitting in DIP-based CT reconstruction
This paper investigates the application of unsupervised learning methods for computed tomography (CT) reconstruction. To motivate our work, we review several existing priors, namely the truncated Gaussian prior, the $l_1$ prior, the total variation prior, and the deep image prior (DIP). We find that DIP outperforms the other three priors in terms of representational capability and visual performance. However, the performance of DIP deteriorates when the number of iterations exceeds a certain threshold due to overfitting. To address this issue, we propose a novel method (MCDIP-ADMM) based on Multi-Code Deep Image Prior and plug-and-play Alternative Direction Method of Multipliers. Specifically, MCDIP utilizes multiple latent codes to generate a series of feature maps at an intermediate layer within a generator model. These maps are then composed with trainable weights, representing the complete image prior. Experimental results demonstrate the superior performance of the proposed MCDIP-ADMM compared to three existing competitors. In the case of parallel beam projection with Gaussian noise, MCDIP-ADMM achieves an average improvement of 4.3 dB over DIP, 1.7 dB over ADMM DIP-WTV, and 1.2 dB over PnP-DIP in terms of PSNR. Similarly, for fan-beam projection with Poisson noise, MCDIP-ADMM achieves an average improvement of 3.09 dB over DIP, 1.86 dB over ADMM DIP-WTV, and 0.84 dB over PnP-DIP in terms of PSNR.
comment: 25 pages
♻ ☆ Point-GCC: Universal Self-supervised 3D Scene Pre-training via Geometry-Color Contrast
Geometry and color information provided by the point clouds are both crucial for 3D scene understanding. Two pieces of information characterize the different aspects of point clouds, but existing methods lack an elaborate design for the discrimination and relevance. Hence we explore a 3D self-supervised paradigm that can better utilize the relations of point cloud information. Specifically, we propose a universal 3D scene pre-training framework via Geometry-Color Contrast (Point-GCC), which aligns geometry and color information using a Siamese network. To take care of actual application tasks, we design (i) hierarchical supervision with point-level contrast and reconstruct and object-level contrast based on the novel deep clustering module to close the gap between pre-training and downstream tasks; (ii) architecture-agnostic backbone to adapt for various downstream models. Benefiting from the object-level representation associated with downstream tasks, Point-GCC can directly evaluate model performance and the result demonstrates the effectiveness of our methods. Transfer learning results on a wide range of tasks also show consistent improvements across all datasets. e.g., new state-of-the-art object detection results on SUN RGB-D and S3DIS datasets. Codes will be released at https://github.com/Asterisci/Point-GCC.
♻ ☆ Interactive Segmentation for COVID-19 Infection Quantification on Longitudinal CT scans
Consistent segmentation of COVID-19 patient's CT scans across multiple time points is essential to assess disease progression and response to therapy accurately. Existing automatic and interactive segmentation models for medical images only use data from a single time point (static). However, valuable segmentation information from previous time points is often not used to aid the segmentation of a patient's follow-up scans. Also, fully automatic segmentation techniques frequently produce results that would need further editing for clinical use. In this work, we propose a new single network model for interactive segmentation that fully utilizes all available past information to refine the segmentation of follow-up scans. In the first segmentation round, our model takes 3D volumes of medical images from two-time points (target and reference) as concatenated slices with the additional reference time point segmentation as a guide to segment the target scan. In subsequent segmentation refinement rounds, user feedback in the form of scribbles that correct the segmentation and the target's previous segmentation results are additionally fed into the model. This ensures that the segmentation information from previous refinement rounds is retained. Experimental results on our in-house multiclass longitudinal COVID-19 dataset show that the proposed model outperforms its static version and can assist in localizing COVID-19 infections in patient's follow-up scans.
comment: 10 pages, 11 figures, 4 tables
♻ ☆ Direct Learning-Based Deep Spiking Neural Networks: A Review
The spiking neural network (SNN), as a promising brain-inspired computational model with binary spike information transmission mechanism, rich spatially-temporal dynamics, and event-driven characteristics, has received extensive attention. However, its intricately discontinuous spike mechanism brings difficulty to the optimization of the deep SNN. Since the surrogate gradient method can greatly mitigate the optimization difficulty and shows great potential in directly training deep SNNs, a variety of direct learning-based deep SNN works have been proposed and achieved satisfying progress in recent years. In this paper, we present a comprehensive survey of these direct learning-based deep SNN works, mainly categorized into accuracy improvement methods, efficiency improvement methods, and temporal dynamics utilization methods. In addition, we also divide these categorizations into finer granularities further to better organize and introduce them. Finally, the challenges and trends that may be faced in future research are prospected.
comment: Accepted by Frontiers in Neuroscience
♻ ☆ Pattern Attention Transformer with Doughnut Kernel
We present in this paper a new architecture, the Pattern Attention Transformer (PAT), that is composed of the new doughnut kernel. Compared with tokens in the NLP field, Transformer in computer vision has the problem of handling the high resolution of pixels in images. In ViT, an image is cut into square-shaped patches. As the follow-up of ViT, Swin Transformer proposes an additional step of shifting to decrease the existence of fixed boundaries, which also incurs 'two connected Swin Transformer blocks' as the minimum unit of the model. Inheriting the patch/window idea, our doughnut kernel enhances the design of patches further. It replaces the line-cut boundaries with two types of areas: sensor and updating, which is based on the comprehension of self-attention (named QKVA grid). The doughnut kernel also brings a new topic about the shape of kernels beyond square. To verify its performance on image classification, PAT is designed with Transformer blocks of regular octagon shape doughnut kernels. Its architecture is lighter: the minimum pattern attention layer is only one for each stage. Under similar complexity of computation, its performances on ImageNet 1K reach higher throughput (+10%) and surpass Swin Transformer (+0.8 acc1).
♻ ☆ When does Privileged Information Explain Away Label Noise? ICML 2023
Leveraging privileged information (PI), or features available during training but not at test time, has recently been shown to be an effective method for addressing label noise. However, the reasons for its effectiveness are not well understood. In this study, we investigate the role played by different properties of the PI in explaining away label noise. Through experiments on multiple datasets with real PI (CIFAR-N/H) and a new large-scale benchmark ImageNet-PI, we find that PI is most helpful when it allows networks to easily distinguish clean from noisy data, while enabling a learning shortcut to memorize the noisy examples. Interestingly, when PI becomes too predictive of the target label, PI methods often perform worse than their no-PI baselines. Based on these findings, we propose several enhancements to the state-of-the-art PI methods and demonstrate the potential of PI as a means of tackling label noise. Finally, we show how we can easily combine the resulting PI approaches with existing no-PI techniques designed to deal with label noise.
comment: Accepted ICML 2023, Honolulu
♻ ☆ Late-Constraint Diffusion Guidance for Controllable Image Synthesis
Diffusion models, either with or without text condition, have demonstrated impressive capability in synthesizing photorealistic images given a few or even no words. These models may not fully satisfy user need, as normal users or artists intend to control the synthesized images with specific guidance, like overall layout, color, structure, object shape, and so on. To adapt diffusion models for controllable image synthesis, several methods have been proposed to incorporate the required conditions as regularization upon the intermediate features of the diffusion denoising network. These methods, known as early-constraint ones in this paper, have difficulties in handling multiple conditions with a single solution. They intend to train separate models for each specific condition, which require much training cost and result in non-generalizable solutions. To address these difficulties, we propose a new approach namely late-constraint: we leave the diffusion networks unchanged, but constrain its output to be aligned with the required conditions. Specifically, we train a lightweight condition adapter to establish the correlation between external conditions and internal representations of diffusion models. During the iterative denoising process, the conditional guidance is sent into corresponding condition adapter to manipulate the sampling process with the established correlation. We further equip the introduced late-constraint strategy with a timestep resampling method and an early stopping technique, which boost the quality of synthesized image meanwhile complying with the guidance. Our method outperforms the existing early-constraint methods and generalizes better to unseen condition. Our code would be available.
♻ ☆ Word separation in continuous sign language using isolated signs and post-processing
. Continuous Sign Language Recognition (CSLR) is a long challenging task in Computer Vision due to the difficulties in detecting the explicit boundaries between the words in a sign sentence. To deal with this challenge, we propose a two-stage model. In the first stage, the predictor model, which includes a combination of CNN, SVD, and LSTM, is trained with the isolated signs. In the second stage, we apply a post-processing algorithm to the Softmax outputs obtained from the first part of the model in order to separate the isolated signs in the continuous signs. While the proposed model is trained on the isolated sign classes with similar frame numbers, it is evaluated on the continuous sign videos with a different frame length per each isolated sign class. Due to the lack of a large dataset, including both the sign sequences and the corresponding isolated signs, two public datasets in Isolated Sign Language Recognition (ISLR), RKS-PERSIANSIGN and ASLLVD, are used for evaluation. Results of the continuous sign videos confirm the efficiency of the proposed model to deal with isolated sign boundaries detection.
♻ ☆ Rethinking Local Perception in Lightweight Vision Transformer
Vision Transformers (ViTs) have been shown to be effective in various vision tasks. However, resizing them to a mobile-friendly size leads to significant performance degradation. Therefore, developing lightweight vision transformers has become a crucial area of research. This paper introduces CloFormer, a lightweight vision transformer that leverages context-aware local enhancement. CloFormer explores the relationship between globally shared weights often used in vanilla convolutional operators and token-specific context-aware weights appearing in attention, then proposes an effective and straightforward module to capture high-frequency local information. In CloFormer, we introduce AttnConv, a convolution operator in attention's style. The proposed AttnConv uses shared weights to aggregate local information and deploys carefully designed context-aware weights to enhance local features. The combination of the AttnConv and vanilla attention which uses pooling to reduce FLOPs in CloFormer enables the model to perceive high-frequency and low-frequency information. Extensive experiments were conducted in image classification, object detection, and semantic segmentation, demonstrating the superiority of CloFormer. The code is available at \url{https://github.com/qhfan/CloFormer}.
♻ ☆ Learning Decorrelated Representations Efficiently Using Fast Fourier Transform CVPR 2023
Barlow Twins and VICReg are self-supervised representation learning models that use regularizers to decorrelate features. Although these models are as effective as conventional representation learning models, their training can be computationally demanding if the dimension d of the projected embeddings is high. As the regularizers are defined in terms of individual elements of a cross-correlation or covariance matrix, computing the loss for n samples takes O(n d^2) time. In this paper, we propose a relaxed decorrelating regularizer that can be computed in O(n d log d) time by Fast Fourier Transform. We also propose an inexpensive technique to mitigate undesirable local minima that develop with the relaxation. The proposed regularizer exhibits accuracy comparable to that of existing regularizers in downstream tasks, whereas their training requires less memory and is faster for large d. The source code is available.
comment: Accepted for CVPR 2023
♻ ☆ Aerial Vision-and-Dialog Navigation ACL 2023
The ability to converse with humans and follow natural language commands is crucial for intelligent unmanned aerial vehicles (a.k.a. drones). It can relieve people's burden of holding a controller all the time, allow multitasking, and make drone control more accessible for people with disabilities or with their hands occupied. To this end, we introduce Aerial Vision-and-Dialog Navigation (AVDN), to navigate a drone via natural language conversation. We build a drone simulator with a continuous photorealistic environment and collect a new AVDN dataset of over 3k recorded navigation trajectories with asynchronous human-human dialogs between commanders and followers. The commander provides initial navigation instruction and further guidance by request, while the follower navigates the drone in the simulator and asks questions when needed. During data collection, followers' attention on the drone's visual observation is also recorded. Based on the AVDN dataset, we study the tasks of aerial navigation from (full) dialog history and propose an effective Human Attention Aided Transformer model (HAA-Transformer), which learns to predict both navigation waypoints and human attention.
comment: Accepted by ACL 2023 Findings
♻ ☆ Multi-Mode Online Knowledge Distillation for Self-Supervised Visual Representation Learning CVPR 2023
Self-supervised learning (SSL) has made remarkable progress in visual representation learning. Some studies combine SSL with knowledge distillation (SSL-KD) to boost the representation learning performance of small models. In this study, we propose a Multi-mode Online Knowledge Distillation method (MOKD) to boost self-supervised visual representation learning. Different from existing SSL-KD methods that transfer knowledge from a static pre-trained teacher to a student, in MOKD, two different models learn collaboratively in a self-supervised manner. Specifically, MOKD consists of two distillation modes: self-distillation and cross-distillation modes. Among them, self-distillation performs self-supervised learning for each model independently, while cross-distillation realizes knowledge interaction between different models. In cross-distillation, a cross-attention feature search strategy is proposed to enhance the semantic feature alignment between different models. As a result, the two models can absorb knowledge from each other to boost their representation learning performance. Extensive experimental results on different backbones and datasets demonstrate that two heterogeneous models can benefit from MOKD and outperform their independently trained baseline. In addition, MOKD also outperforms existing SSL-KD methods for both the student and teacher models.
comment: Accepted by CVPR 2023
♻ ☆ Oscillation-free Quantization for Low-bit Vision Transformers
Weight oscillation is an undesirable side effect of quantization-aware training, in which quantized weights frequently jump between two quantized levels, resulting in training instability and a sub-optimal final model. We discover that the learnable scaling factor, a widely-used $\textit{de facto}$ setting in quantization aggravates weight oscillation. In this study, we investigate the connection between the learnable scaling factor and quantized weight oscillation and use ViT as a case driver to illustrate the findings and remedies. In addition, we also found that the interdependence between quantized weights in $\textit{query}$ and $\textit{key}$ of a self-attention layer makes ViT vulnerable to oscillation. We, therefore, propose three techniques accordingly: statistical weight quantization ($\rm StatsQ$) to improve quantization robustness compared to the prevalent learnable-scale-based method; confidence-guided annealing ($\rm CGA$) that freezes the weights with $\textit{high confidence}$ and calms the oscillating weights; and $\textit{query}$-$\textit{key}$ reparameterization ($\rm QKR$) to resolve the query-key intertwined oscillation and mitigate the resulting gradient misestimation. Extensive experiments demonstrate that these proposed techniques successfully abate weight oscillation and consistently achieve substantial accuracy improvement on ImageNet. Specifically, our 2-bit DeiT-T/DeiT-S algorithms outperform the previous state-of-the-art by 9.8% and 7.7%, respectively. Code and models are available at: \url{https://github.com/nbasyl/OFQ}.
comment: 11 pages, 11 figures
♻ ☆ Implementation of a Blind navigation method in outdoors/indoors areas
According to WHO statistics, the number of visually impaired people is increasing annually. One of the most critical necessities for visually impaired people is the ability to navigate safely. This paper proposes a navigation system based on the visual slam and Yolo algorithm using monocular cameras. The proposed system consists of three steps: obstacle distance estimation, path deviation detection, and next-step prediction. Using the ORB-SLAM algorithm, the proposed method creates a map from a predefined route and guides the users to stay on the route while notifying them if they deviate from it. Additionally, the system utilizes the YOLO algorithm to detect obstacles along the route and alert the user. The experimental results, obtained by using a laptop camera, show that the proposed system can run in 30 frame per second while guiding the user within predefined routes of 11 meters in indoors and outdoors. The accuracy of the positioning system is 8cm, and the system notifies the users if they deviate from the predefined route by more than 60 cm.
comment: 14 pages, 6 figures and 6 tables
♻ ☆ Crafting Training Degradation Distribution for the Accuracy-Generalization Trade-off in Real-World Super-Resolution ICML 2023
Super-resolution (SR) techniques designed for real-world applications commonly encounter two primary challenges: generalization performance and restoration accuracy. We demonstrate that when methods are trained using complex, large-range degradations to enhance generalization, a decline in accuracy is inevitable. However, since the degradation in a certain real-world applications typically exhibits a limited variation range, it becomes feasible to strike a trade-off between generalization performance and testing accuracy within this scope. In this work, we introduce a novel approach to craft training degradation distributions using a small set of reference images. Our strategy is founded upon the binned representation of the degradation space and the Fr\'echet distance between degradation distributions. Our results indicate that the proposed technique significantly improves the performance of test images while preserving generalization capabilities in real-world applications.
comment: This paper has been accepted to ICML 2023
♻ ☆ InDL: A New Datasets and Benchmark for In-Diagram Logic Interpreting based on Visual Illusion
This paper introduces a novel approach to evaluating deep learning models' capacity for in-diagram logic interpretation. Leveraging the intriguing realm of visual illusions, we establish a unique dataset, InDL, designed to rigorously test and benchmark these models. Deep learning has witnessed remarkable progress in domains such as computer vision and natural language processing. However, models often stumble in tasks requiring logical reasoning due to their inherent 'black box' characteristics, which obscure the decision-making process. Our work presents a new lens to understand these models better by focusing on their handling of visual illusions -- a complex interplay of perception and logic. We utilize six classic geometric optical illusions to create a comparative framework between human and machine visual perception. This methodology offers a quantifiable measure to rank models, elucidating potential weaknesses and providing actionable insights for model improvements. Our experimental results affirm the efficacy of our benchmarking strategy, demonstrating its ability to effectively rank models based on their logic interpretation ability. As part of our commitment to reproducible research, the source code and datasets will be made publicly available here: \href{https://github.com/rabbit-magic-wh/InDL}{https://github.com/rabbit-magic-wh/InDL}.
♻ ☆ Spontaneous symmetry breaking in generative diffusion models
Generative diffusion models have recently emerged as a leading approach for generating high-dimensional data. In this paper, we show that the dynamics of these models exhibit a spontaneous symmetry breaking that divides the generative dynamics into two distinct phases: 1) A linear steady-state dynamics around a central fixed-point and 2) an attractor dynamics directed towards the data manifold. These two "phases" are separated by the change in stability of the central fixed-point, with the resulting window of instability being responsible for the diversity of the generated samples. Using both theoretical and empirical evidence, we show that an accurate simulation of the early dynamics does not significantly contribute to the final generation, since early fluctuations are reverted to the central fixed point. To leverage this insight, we propose a Gaussian late initialization scheme, which significantly improves model performance, achieving up to 3x FID improvements on fast samplers, while also increasing sample diversity (e.g., racial composition of generated CelebA images). Our work offers a new way to understand the generative dynamics of diffusion models that has the potential to bring about higher performance and less biased fast-samplers.
comment: typo corrected
♻ ☆ Large Language Models are Frame-level Directors for Zero-shot Text-to-Video Generation
In the paradigm of AI-generated content (AIGC), there has been increasing attention in extending pre-trained text-to-image (T2I) models to text-to-video (T2V) generation. Despite their effectiveness, these frameworks face challenges in maintaining consistent narratives and handling rapid shifts in scene composition or object placement from a single user prompt. This paper introduces a new framework, dubbed DirecT2V, which leverages instruction-tuned large language models (LLMs) to generate frame-by-frame descriptions from a single abstract user prompt. DirecT2V utilizes LLM directors to divide user inputs into separate prompts for each frame, enabling the inclusion of time-varying content and facilitating consistent video generation. To maintain temporal consistency and prevent object collapse, we propose a novel value mapping method and dual-softmax filtering. Extensive experimental results validate the effectiveness of the DirecT2V framework in producing visually coherent and consistent videos from abstract user prompts, addressing the challenges of zero-shot video generation.
comment: The code and demo will be available at https://github.com/KU-CVLAB/DirecT2V
♻ ☆ Super-CLEVR: A Virtual Benchmark to Diagnose Domain Robustness in Visual Reasoning CVPR 2023
Visual Question Answering (VQA) models often perform poorly on out-of-distribution data and struggle on domain generalization. Due to the multi-modal nature of this task, multiple factors of variation are intertwined, making generalization difficult to analyze. This motivates us to introduce a virtual benchmark, Super-CLEVR, where different factors in VQA domain shifts can be isolated in order that their effects can be studied independently. Four factors are considered: visual complexity, question redundancy, concept distribution and concept compositionality. With controllably generated data, Super-CLEVR enables us to test VQA methods in situations where the test data differs from the training data along each of these axes. We study four existing methods, including two neural symbolic methods NSCL and NSVQA, and two non-symbolic methods FiLM and mDETR; and our proposed method, probabilistic NSVQA (P-NSVQA), which extends NSVQA with uncertainty reasoning. P-NSVQA outperforms other methods on three of the four domain shift factors. Our results suggest that disentangling reasoning and perception, combined with probabilistic uncertainty, form a strong VQA model that is more robust to domain shifts. The dataset and code are released at https://github.com/Lizw14/Super-CLEVR.
comment: Published in CVPR 2023 as Highlight. Data and code are released at https://github.com/Lizw14/Super-CLEVR
♻ ☆ HyperMODEST: Self-Supervised 3D Object Detection with Confidence Score Filtering
Current LiDAR-based 3D object detectors for autonomous driving are almost entirely trained on human-annotated data collected in specific geographical domains with specific sensor setups, making it difficult to adapt to a different domain. MODEST is the first work to train 3D object detectors without any labels. Our work, HyperMODEST, proposes a universal method implemented on top of MODEST that can largely accelerate the self-training process and does not require tuning on a specific dataset. We filter intermediate pseudo-labels used for data augmentation with low confidence scores. On the nuScenes dataset, we observe a significant improvement of 1.6% in AP BEV in 0-80m range at IoU=0.25 and an improvement of 1.7% in AP BEV in 0-80m range at IoU=0.5 while only using one-fifth of the training time in the original approach by MODEST. On the Lyft dataset, we also observe an improvement over the baseline during the first round of iterative self-training. We explore the trade-off between high precision and high recall in the early stage of the self-training process by comparing our proposed method with two other score filtering methods: confidence score filtering for pseudo-labels with and without static label retention. The code and models of this work are available at https://github.com/TRAILab/HyperMODEST
comment: Accepted in CRV (Conference on Robots and Vision) 2023
♻ ☆ From Malware Samples to Fractal Images: A New Paradigm for Classification. (Version 2.0, Previous version paper name: Have you ever seen malware?)
To date, a large number of research papers have been written on the classification of malware, its identification, classification into different families and the distinction between malware and goodware. These works have been based on captured malware samples and have attempted to analyse malware and goodware using various techniques, including techniques from the field of artificial intelligence. For example, neural networks have played a significant role in these classification methods. Some of this work also deals with analysing malware using its visualisation. These works usually convert malware samples capturing the structure of malware into image structures, which are then the object of image processing. In this paper, we propose a very unconventional and novel approach to malware visualisation based on dynamic behaviour analysis, with the idea that the images, which are visually very interesting, are then used to classify malware concerning goodware. Our approach opens an extensive topic for future discussion and provides many new directions for research in malware analysis and classification, as discussed in conclusion. The results of the presented experiments are based on a database of 6 589 997 goodware, 827 853 potentially unwanted applications and 4 174 203 malware samples provided by ESET and selected experimental data (images, generating polynomial formulas and software generating images) are available on GitHub for interested readers. Thus, this paper is not a comprehensive compact study that reports the results obtained from comparative experiments but rather attempts to show a new direction in the field of visualisation with possible applications in malware analysis.
comment: This paper is under review; the section describing conversion from malware structure to fractal figure is temporarily erased here to protect our idea. It will be replaced by a full version when accepted
♻ ☆ CHAD: Charlotte Anomaly Dataset
In recent years, we have seen a significant interest in data-driven deep learning approaches for video anomaly detection, where an algorithm must determine if specific frames of a video contain abnormal behaviors. However, video anomaly detection is particularly context-specific, and the availability of representative datasets heavily limits real-world accuracy. Additionally, the metrics currently reported by most state-of-the-art methods often do not reflect how well the model will perform in real-world scenarios. In this article, we present the Charlotte Anomaly Dataset (CHAD). CHAD is a high-resolution, multi-camera anomaly dataset in a commercial parking lot setting. In addition to frame-level anomaly labels, CHAD is the first anomaly dataset to include bounding box, identity, and pose annotations for each actor. This is especially beneficial for skeleton-based anomaly detection, which is useful for its lower computational demand in real-world settings. CHAD is also the first anomaly dataset to contain multiple views of the same scene. With four camera views and over 1.15 million frames, CHAD is the largest fully annotated anomaly detection dataset including person annotations, collected from continuous video streams from stationary cameras for smart video surveillance applications. To demonstrate the efficacy of CHAD for training and evaluation, we benchmark two state-of-the-art skeleton-based anomaly detection algorithms on CHAD and provide comprehensive analysis, including both quantitative results and qualitative examination. The dataset is available at https://github.com/TeCSAR-UNCC/CHAD.
♻ ☆ Generating Adversarial Examples with Task Oriented Multi-Objective Optimization
Deep learning models, even the-state-of-the-art ones, are highly vulnerable to adversarial examples. Adversarial training is one of the most efficient methods to improve the model's robustness. The key factor for the success of adversarial training is the capability to generate qualified and divergent adversarial examples which satisfy some objectives/goals (e.g., finding adversarial examples that maximize the model losses for simultaneously attacking multiple models). Therefore, multi-objective optimization (MOO) is a natural tool for adversarial example generation to achieve multiple objectives/goals simultaneously. However, we observe that a naive application of MOO tends to maximize all objectives/goals equally, without caring if an objective/goal has been achieved yet. This leads to useless effort to further improve the goal-achieved tasks, while putting less focus on the goal-unachieved tasks. In this paper, we propose \emph{Task Oriented MOO} to address this issue, in the context where we can explicitly define the goal achievement for a task. Our principle is to only maintain the goal-achieved tasks, while letting the optimizer spend more effort on improving the goal-unachieved tasks. We conduct comprehensive experiments for our Task Oriented MOO on various adversarial example generation schemes. The experimental results firmly demonstrate the merit of our proposed approach. Our code is available at \url{https://github.com/tuananhbui89/TAMOO}.
Information Retrieval 21
☆ AMR4NLI: Interpretable and robust NLI measures from semantic graphs
The task of natural language inference (NLI) asks whether a given premise (expressed in NL) entails a given NL hypothesis. NLI benchmarks contain human ratings of entailment, but the meaning relationships driving these ratings are not formalized. Can the underlying sentence pair relationships be made more explicit in an interpretable yet robust fashion? We compare semantic structures to represent premise and hypothesis, including sets of contextualized embeddings and semantic graphs (Abstract Meaning Representations), and measure whether the hypothesis is a semantic substructure of the premise, utilizing interpretable metrics. Our evaluation on three English benchmarks finds value in both contextualized embeddings and semantic graphs; moreover, they provide complementary signals, and can be leveraged together in a hybrid model.
comment: International Conference on Computational Semantics (IWCS 2023)
☆ ACLM: A Selective-Denoising based Generative Data Augmentation Approach for Low-Resource Complex NER ACL 2023
Complex Named Entity Recognition (NER) is the task of detecting linguistically complex named entities in low-context text. In this paper, we present ACLM Attention-map aware keyword selection for Conditional Language Model fine-tuning), a novel data augmentation approach based on conditional generation to address the data scarcity problem in low-resource complex NER. ACLM alleviates the context-entity mismatch issue, a problem existing NER data augmentation techniques suffer from and often generates incoherent augmentations by placing complex named entities in the wrong context. ACLM builds on BART and is optimized on a novel text reconstruction or denoising task - we use selective masking (aided by attention maps) to retain the named entities and certain keywords in the input sentence that provide contextually relevant additional knowledge or hints about the named entities. Compared with other data augmentation strategies, ACLM can generate more diverse and coherent augmentations preserving the true word sense of complex entities in the sentence. We demonstrate the effectiveness of ACLM both qualitatively and quantitatively on monolingual, cross-lingual, and multilingual complex NER across various low-resource settings. ACLM outperforms all our neural baselines by a significant margin (1%-36%). In addition, we demonstrate the application of ACLM to other domains that suffer from data scarcity (e.g., biomedical). In practice, ACLM generates more effective and factual augmentations for these domains than prior methods. Code: https://github.com/Sreyan88/ACLM
comment: ACL 2023 Main Conference
☆ SpotTarget: Rethinking the Effect of Target Edges for Link Prediction in Graph Neural Networks
Graph Neural Networks (GNNs) have demonstrated promising outcomes across various tasks, including node classification and link prediction. Despite their remarkable success in various high-impact applications, we have identified three common pitfalls in message passing for link prediction. Particularly, in prevalent GNN frameworks (e.g., DGL and PyTorch-Geometric), the target edges (i.e., the edges being predicted) consistently exist as message passing edges in the graph during training. Consequently, this results in overfitting and distribution shift, both of which adversely impact the generalizability to test the target edges. Additionally, during test time, the failure to exclude the test target edges leads to implicit test leakage caused by neighborhood aggregation. In this paper, we analyze these three pitfalls and investigate the impact of including or excluding target edges on the performance of nodes with varying degrees during training and test phases. Our theoretical and empirical analysis demonstrates that low-degree nodes are more susceptible to these pitfalls. These pitfalls can have detrimental consequences when GNNs are implemented in production systems. To systematically address these pitfalls, we propose SpotTarget, an effective and efficient GNN training framework. During training, SpotTarget leverages our insight regarding low-degree nodes and excludes train target edges connected to at least one low-degree node. During test time, it emulates real-world scenarios of GNN usage in production and excludes all test target edges. Our experiments conducted on diverse real-world datasets, demonstrate that SpotTarget significantly enhances GNNs, achieving up to a 15x increase in accuracy in sparse graphs. Furthermore, SpotTarget consistently and dramatically improves the performance for low-degree nodes in dense graphs.
comment: 8 pages, 2 page appendix
☆ Topic-Guided Sampling For Data-Efficient Multi-Domain Stance Detection ACL 2023
Stance Detection is concerned with identifying the attitudes expressed by an author towards a target of interest. This task spans a variety of domains ranging from social media opinion identification to detecting the stance for a legal claim. However, the framing of the task varies within these domains, in terms of the data collection protocol, the label dictionary and the number of available annotations. Furthermore, these stance annotations are significantly imbalanced on a per-topic and inter-topic basis. These make multi-domain stance detection a challenging task, requiring standardization and domain adaptation. To overcome this challenge, we propose $\textbf{T}$opic $\textbf{E}$fficient $\textbf{St}$anc$\textbf{E}$ $\textbf{D}$etection (TESTED), consisting of a topic-guided diversity sampling technique and a contrastive objective that is used for fine-tuning a stance classifier. We evaluate the method on an existing benchmark of $16$ datasets with in-domain, i.e. all topics seen and out-of-domain, i.e. unseen topics, experiments. The results show that our method outperforms the state-of-the-art with an average of $3.5$ F1 points increase in-domain, and is more generalizable with an averaged increase of $10.2$ F1 on out-of-domain evaluation while using $\leq10\%$ of the training data. We show that our sampling technique mitigates both inter- and per-topic class imbalances. Finally, our analysis demonstrates that the contrastive learning objective allows the model a more pronounced segmentation of samples with varying labels.
comment: ACL 2023 (Oral)
☆ End-to-End Document Classification and Key Information Extraction using Assignment Optimization
We propose end-to-end document classification and key information extraction (KIE) for automating document processing in forms. Through accurate document classification we harness known information from templates to enhance KIE from forms. We use text and layout encoding with a cosine similarity measure to classify visually-similar documents. We then demonstrate a novel application of mixed integer programming by using assignment optimization to extract key information from documents. Our approach is validated on an in-house dataset of noisy scanned forms. The best performing document classification approach achieved 0.97 f1 score. A mean f1 score of 0.94 for the KIE task suggests there is significant potential in applying optimization techniques. Abation results show that the method relies on document preprocessing techniques to mitigate Type II errors and achieve optimal performance.
comment: 10 pages, 5 figures
☆ Class Anchor Margin Loss for Content-Based Image Retrieval
The performance of neural networks in content-based image retrieval (CBIR) is highly influenced by the chosen loss (objective) function. The majority of objective functions for neural models can be divided into metric learning and statistical learning. Metric learning approaches require a pair mining strategy that often lacks efficiency, while statistical learning approaches are not generating highly compact features due to their indirect feature optimization. To this end, we propose a novel repeller-attractor loss that falls in the metric learning paradigm, yet directly optimizes for the L2 metric without the need of generating pairs. Our loss is formed of three components. One leading objective ensures that the learned features are attracted to each designated learnable class anchor. The second loss component regulates the anchors and forces them to be separable by a margin, while the third objective ensures that the anchors do not collapse to zero. Furthermore, we develop a more efficient two-stage retrieval system by harnessing the learned class anchors during the first stage of the retrieval process, eliminating the need of comparing the query with every image in the database. We establish a set of four datasets (CIFAR-100, Food-101, SVHN, and Tiny ImageNet) and evaluate the proposed objective in the context of few-shot and full-set training on the CBIR task, by using both convolutional and transformer architectures. Compared to existing objective functions, our empirical evidence shows that the proposed objective is generating superior and more consistent results.
☆ End-to-end Knowledge Retrieval with Multi-modal Queries ACL 2023
We investigate knowledge retrieval with multi-modal queries, i.e. queries containing information split across image and text inputs, a challenging task that differs from previous work on cross-modal retrieval. We curate a new dataset called ReMuQ for benchmarking progress on this task. ReMuQ requires a system to retrieve knowledge from a large corpus by integrating contents from both text and image queries. We introduce a retriever model ``ReViz'' that can directly process input text and images to retrieve relevant knowledge in an end-to-end fashion without being dependent on intermediate modules such as object detectors or caption generators. We introduce a new pretraining task that is effective for learning knowledge retrieval with multimodal queries and also improves performance on downstream tasks. We demonstrate superior performance in retrieval on two datasets (ReMuQ and OK-VQA) under zero-shot settings as well as further improvements when finetuned on these datasets.
comment: ACL 2023
☆ A Survey on Fairness-aware Recommender Systems
As information filtering services, recommender systems have extremely enriched our daily life by providing personalized suggestions and facilitating people in decision-making, which makes them vital and indispensable to human society in the information era. However, as people become more dependent on them, recent studies show that recommender systems potentially own unintentional impacts on society and individuals because of their unfairness (e.g., gender discrimination in job recommendations). To develop trustworthy services, it is crucial to devise fairness-aware recommender systems that can mitigate these bias issues. In this survey, we summarise existing methodologies and practices of fairness in recommender systems. Firstly, we present concepts of fairness in different recommendation scenarios, comprehensively categorize current advances, and introduce typical methods to promote fairness in different stages of recommender systems. Next, after introducing datasets and evaluation metrics applied to assess the fairness of recommender systems, we will delve into the significant influence that fairness-aware recommender systems exert on real-world industrial applications. Subsequently, we highlight the connection between fairness and other principles of trustworthy recommender systems, aiming to consider trustworthiness principles holistically while advocating for fairness. Finally, we summarize this review, spotlighting promising opportunities in comprehending concepts, frameworks, the balance between accuracy and fairness, and the ties with trustworthiness, with the ultimate goal of fostering the development of fairness-aware recommender systems.
comment: 27 pages, 9 figures
☆ Explicit Feature Interaction-aware Uplift Network for Online Marketing KDD 2023
As a key component in online marketing, uplift modeling aims to accurately capture the degree to which different treatments motivate different users, such as coupons or discounts, also known as the estimation of individual treatment effect (ITE). In an actual business scenario, the options for treatment may be numerous and complex, and there may be correlations between different treatments. In addition, each marketing instance may also have rich user and contextual features. However, existing methods still fall short in both fully exploiting treatment information and mining features that are sensitive to a particular treatment. In this paper, we propose an explicit feature interaction-aware uplift network (EFIN) to address these two problems. Our EFIN includes four customized modules: 1) a feature encoding module encodes not only the user and contextual features, but also the treatment features; 2) a self-interaction module aims to accurately model the user's natural response with all but the treatment features; 3) a treatment-aware interaction module accurately models the degree to which a particular treatment motivates a user through interactions between the treatment features and other features, i.e., ITE; and 4) an intervention constraint module is used to balance the ITE distribution of users between the control and treatment groups so that the model would still achieve a accurate uplift ranking on data collected from a non-random intervention marketing scenario. We conduct extensive experiments on two public datasets and one product dataset to verify the effectiveness of our EFIN. In addition, our EFIN has been deployed in a credit card bill payment scenario of a large online financial platform with a significant improvement.
comment: Accepted by SIGKDD 2023 Applied Data Science Track
☆ A Multi-Modal Latent-Features based Service Recommendation System for the Social Internet of Things
The Social Internet of Things (SIoT), is revolutionizing how we interact with our everyday lives. By adding the social dimension to connecting devices, the SIoT has the potential to drastically change the way we interact with smart devices. This connected infrastructure allows for unprecedented levels of convenience, automation, and access to information, allowing us to do more with less effort. However, this revolutionary new technology also brings an eager need for service recommendation systems. As the SIoT grows in scope and complexity, it becomes increasingly important for businesses and individuals, and SIoT objects alike to have reliable sources for products, services, and information that are tailored to their specific needs. Few works have been proposed to provide service recommendations for SIoT environments. However, these efforts have been confined to only focusing on modeling user-item interactions using contextual information, devices' SIoT relationships, and correlation social groups but these schemes do not account for latent semantic item-item structures underlying the sparse multi-modal contents in SIoT environment. In this paper, we propose a latent-based SIoT recommendation system that learns item-item structures and aggregates multiple modalities to obtain latent item graphs which are then used in graph convolutions to inject high-order affinities into item representations. Experiments showed that the proposed recommendation system outperformed state-of-the-art SIoT recommendation methods and validated its efficacy at mining latent relationships from multi-modal features.
☆ TimelineQA: A Benchmark for Question Answering over Timelines
Lifelogs are descriptions of experiences that a person had during their life. Lifelogs are created by fusing data from the multitude of digital services, such as online photos, maps, shopping and content streaming services. Question answering over lifelogs can offer personal assistants a critical resource when they try to provide advice in context. However, obtaining answers to questions over lifelogs is beyond the current state of the art of question answering techniques for a variety of reasons, the most pronounced of which is that lifelogs combine free text with some degree of structure such as temporal and geographical information. We create and publicly release TimelineQA1, a benchmark for accelerating progress on querying lifelogs. TimelineQA generates lifelogs of imaginary people. The episodes in the lifelog range from major life episodes such as high school graduation to those that occur on a daily basis such as going for a run. We describe a set of experiments on TimelineQA with several state-of-the-art QA models. Our experiments reveal that for atomic queries, an extractive QA system significantly out-performs a state-of-the-art retrieval-augmented QA system. For multi-hop queries involving aggregates, we show that the best result is obtained with a state-of-the-art table QA technique, assuming the ground truth set of episodes for deriving the answer is available.
☆ Cook-Gen: Robust Generative Modeling of Cooking Actions from Recipes
As people become more aware of their food choices, food computation models have become increasingly popular in assisting people in maintaining healthy eating habits. For example, food recommendation systems analyze recipe instructions to assess nutritional contents and provide recipe recommendations. The recent and remarkable successes of generative AI methods, such as auto-regressive large language models, can lead to robust methods for a more comprehensive understanding of recipes for healthy food recommendations beyond surface-level nutrition content assessments. In this study, we explore the use of generative AI methods to extend current food computation models, primarily involving the analysis of nutrition and ingredients, to also incorporate cooking actions (e.g., add salt, fry the meat, boil the vegetables, etc.). Cooking actions are notoriously hard to model using statistical learning methods due to irregular data patterns - significantly varying natural language descriptions for the same action (e.g., marinate the meat vs. marinate the meat and leave overnight) and infrequently occurring patterns (e.g., add salt occurs far more frequently than marinating the meat). The prototypical approach to handling irregular data patterns is to increase the volume of data that the model ingests by orders of magnitude. Unfortunately, in the cooking domain, these problems are further compounded with larger data volumes presenting a unique challenge that is not easily handled by simply scaling up. In this work, we propose novel aggregation-based generative AI methods, Cook-Gen, that reliably generate cooking actions from recipes, despite difficulties with irregular data patterns, while also outperforming Large Language Models and other strong baselines.
☆ Pairwise Ranking Losses of Click-Through Rates Prediction for Welfare Maximization in Ad Auctions
We study the design of loss functions for click-through rates (CTR) to optimize (social) welfare in advertising auctions. Existing works either only focus on CTR predictions without consideration of business objectives (e.g., welfare) in auctions or assume that the distribution over the participants' expected cost-per-impression (eCPM) is known a priori, then use various additional assumptions on the parametric form of the distribution to derive loss functions for predicting CTRs. In this work, we bring back the welfare objectives of ad auctions into CTR predictions and propose a novel weighted rankloss to train the CTR model. Compared to existing literature, our approach provides a provable guarantee on welfare but without assumptions on the eCPMs' distribution while also avoiding the intractability of naively applying existing learning-to-rank methods. Further, we propose a theoretically justifiable technique for calibrating the losses using labels generated from a teacher network, only assuming that the teacher network has bounded $\ell_2$ generalization error. Finally, we demonstrate the advantages of the proposed loss on synthetic and real-world data.
comment: 25 pages, 6 figures
☆ Task Relation-aware Continual User Representation Learning KDD 2023
User modeling, which learns to represent users into a low-dimensional representation space based on their past behaviors, got a surge of interest from the industry for providing personalized services to users. Previous efforts in user modeling mainly focus on learning a task-specific user representation that is designed for a single task. However, since learning task-specific user representations for every task is infeasible, recent studies introduce the concept of universal user representation, which is a more generalized representation of a user that is relevant to a variety of tasks. Despite their effectiveness, existing approaches for learning universal user representations are impractical in real-world applications due to the data requirement, catastrophic forgetting and the limited learning capability for continually added tasks. In this paper, we propose a novel continual user representation learning method, called TERACON, whose learning capability is not limited as the number of learned tasks increases while capturing the relationship between the tasks. The main idea is to introduce an embedding for each task, i.e., task embedding, which is utilized to generate task-specific soft masks that not only allow the entire model parameters to be updated until the end of training sequence, but also facilitate the relationship between the tasks to be captured. Moreover, we introduce a novel knowledge retention module with pseudo-labeling strategy that successfully alleviates the long-standing problem of continual learning, i.e., catastrophic forgetting. Extensive experiments on public and proprietary real-world datasets demonstrate the superiority and practicality of TERACON. Our code is available at https://github.com/Sein-Kim/TERACON.
comment: KDD 2023
♻ ☆ Efficient Bi-Level Optimization for Recommendation Denoising
The acquisition of explicit user feedback (e.g., ratings) in real-world recommender systems is often hindered by the need for active user involvement. To mitigate this issue, implicit feedback (e.g., clicks) generated during user browsing is exploited as a viable substitute. However, implicit feedback possesses a high degree of noise, which significantly undermines recommendation quality. While many methods have been proposed to address this issue by assigning varying weights to implicit feedback, two shortcomings persist: (1) the weight calculation in these methods is iteration-independent, without considering the influence of weights in previous iterations, and (2) the weight calculation often relies on prior knowledge, which may not always be readily available or universally applicable. To overcome these two limitations, we model recommendation denoising as a bi-level optimization problem. The inner optimization aims to derive an effective model for the recommendation, as well as guiding the weight determination, thereby eliminating the need for prior knowledge. The outer optimization leverages gradients of the inner optimization and adjusts the weights in a manner considering the impact of previous weights. To efficiently solve this bi-level optimization problem, we employ a weight generator to avoid the storage of weights and a one-step gradient-matching-based loss to significantly reduce computational time. The experimental results on three benchmark datasets demonstrate that our proposed approach outperforms both state-of-the-art general and denoising recommendation models. The code is available at https://github.com/CoderWZW/BOD.
comment: 11pages, 5 figures, 6 tables
♻ ☆ Explaining Recommendation System Using Counterfactual Textual Explanations
Currently, there is a significant amount of research being conducted in the field of artificial intelligence to improve the explainability and interpretability of deep learning models. It is found that if end-users understand the reason for the production of some output, it is easier to trust the system. Recommender systems are one example of systems that great efforts have been conducted to make their output more explainable. One method for producing a more explainable output is using counterfactual reasoning, which involves altering minimal features to generate a counterfactual item that results in changing the output of the system. This process allows the identification of input features that have a significant impact on the desired output, leading to effective explanations. In this paper, we present a method for generating counterfactual explanations for both tabular and textual features. We evaluated the performance of our proposed method on three real-world datasets and demonstrated a +5\% improvement on finding effective features (based on model-based measures) compared to the baseline method.
♻ ☆ How to Unleash the Power of Large Language Models for Few-shot Relation Extraction? ACL 2023
Scaling language models have revolutionized widespread NLP tasks, yet little comprehensively explored few-shot relation extraction with large language models. In this paper, we investigate principal methodologies, in-context learning and data generation, for few-shot relation extraction via GPT-3.5 through exhaustive experiments. To enhance few-shot performance, we further propose task-related instructions and schema-constrained data generation. We observe that in-context learning can achieve performance on par with previous prompt learning approaches, and data generation with the large language model can boost previous solutions to obtain new state-of-the-art few-shot results on four widely-studied relation extraction datasets. We hope our work can inspire future research for the capabilities of large language models in few-shot relation extraction. Code is available in https://github.com/zjunlp/DeepKE/tree/main/example/llm.
comment: SustaiNLP Workshop@ACL 2023
♻ ☆ Graph Masked Autoencoder for Sequential Recommendation SIGIR 2023
While some powerful neural network architectures (e.g., Transformer, Graph Neural Networks) have achieved improved performance in sequential recommendation with high-order item dependency modeling, they may suffer from poor representation capability in label scarcity scenarios. To address the issue of insufficient labels, Contrastive Learning (CL) has attracted much attention in recent methods to perform data augmentation through embedding contrasting for self-supervision. However, due to the hand-crafted property of their contrastive view generation strategies, existing CL-enhanced models i) can hardly yield consistent performance on diverse sequential recommendation tasks; ii) may not be immune to user behavior data noise. In light of this, we propose a simple yet effective Graph Masked AutoEncoder-enhanced sequential Recommender system (MAERec) that adaptively and dynamically distills global item transitional information for self-supervised augmentation. It naturally avoids the above issue of heavy reliance on constructing high-quality embedding contrastive views. Instead, an adaptive data reconstruction paradigm is designed to be integrated with the long-range item dependency modeling, for informative augmentation in sequential recommendation. Extensive experiments demonstrate that our method significantly outperforms state-of-the-art baseline models and can learn more accurate representations against data noise and sparsity. Our implemented model code is available at https://github.com/HKUDS/MAERec.
comment: This paper has been published as a full paper at SIGIR 2023
♻ ☆ Criteria Tell You More than Ratings: Criteria Preference-Aware Light Graph Convolution for Effective Multi-Criteria Recommendation KDD
The multi-criteria (MC) recommender system, which leverages MC rating information in a wide range of e-commerce areas, is ubiquitous nowadays. Surprisingly, although graph neural networks (GNNs) have been widely applied to develop various recommender systems due to GNN's high expressive capability in learning graph representations, it has been still unexplored how to design MC recommender systems with GNNs. In light of this, we make the first attempt towards designing a GNN-aided MC recommender system. Specifically, rather than straightforwardly adopting existing GNN-based recommendation methods, we devise a novel criteria preference-aware light graph convolution CPA-LGC method, which is capable of precisely capturing the criteria preference of users as well as the collaborative signal in complex high-order connectivities. To this end, we first construct an MC expansion graph that transforms user--item MC ratings into an expanded bipartite graph to potentially learn from the collaborative signal in MC ratings. Next, to strengthen the capability of criteria preference awareness, CPA-LGC incorporates newly characterized embeddings, including user-specific criteria-preference embeddings and item-specific criterion embeddings, into our graph convolution model. Through comprehensive evaluations using four real-world datasets, we demonstrate (a) the superiority over benchmark MC recommendation methods and benchmark recommendation methods using GNNs with tremendous gains, (b) the effectiveness of core components in CPA-LGC, and (c) the computational efficiency.
comment: 12 pages, 10 figures, 5 tables; 29th ACM SIGKDD Conference on Knowledge Discovery & Data (KDD 2023) (to appear) (Please cite our conference version.)
♻ ☆ A Survey on Large Language Models for Recommendation
Large Language Models (LLMs) have emerged as powerful tools in the field of Natural Language Processing (NLP) and have recently gained significant attention in the domain of Recommendation Systems (RS). These models, trained on massive amounts of data using self-supervised learning, have demonstrated remarkable success in learning universal representations and have the potential to enhance various aspects of recommendation systems by some effective transfer techniques such as fine-tuning and prompt tuning, and so on. The crucial aspect of harnessing the power of language models in enhancing recommendation quality is the utilization of their high-quality representations of textual features and their extensive coverage of external knowledge to establish correlations between items and users. To provide a comprehensive understanding of the existing LLM-based recommendation systems, this survey presents a taxonomy that categorizes these models into two major paradigms, respectively Discriminative LLM for Recommendation (DLLM4Rec) and Generative LLM for Recommendation (GLLM4Rec), with the latter being systematically sorted out for the first time. Furthermore, we systematically review and analyze existing LLM-based recommendation systems within each paradigm, providing insights into their methodologies, techniques, and performance. Additionally, we identify key challenges and several valuable findings to provide researchers and practitioners with inspiration. We have also created a GitHub repository to index relevant papers on LLMs for recommendation, https://github.com/WLiK/LLM4Rec.
comment: 10 pages, 3 figures
♻ ☆ Quick Dense Retrievers Consume KALE: Post Training Kullback Leibler Alignment of Embeddings for Asymmetrical dual encoders ACL 2023
In this paper, we consider the problem of improving the inference latency of language model-based dense retrieval systems by introducing structural compression and model size asymmetry between the context and query encoders. First, we investigate the impact of pre and post-training compression on the MSMARCO, Natural Questions, TriviaQA, SQUAD, and SCIFACT, finding that asymmetry in the dual encoders in dense retrieval can lead to improved inference efficiency. Knowing this, we introduce Kullback Leibler Alignment of Embeddings (KALE), an efficient and accurate method for increasing the inference efficiency of dense retrieval methods by pruning and aligning the query encoder after training. Specifically, KALE extends traditional Knowledge Distillation after bi-encoder training, allowing for effective query encoder compression without full retraining or index generation. Using KALE and asymmetric training, we can generate models which exceed the performance of DistilBERT despite having 3x faster inference.
comment: SustaiNLP2023 @ ACL 2023, 8 pages, 4 figures, 30 tables
Machine Learning 150
☆ Hiera: A Hierarchical Vision Transformer without the Bells-and-Whistles ICML 2023
Modern hierarchical vision transformers have added several vision-specific components in the pursuit of supervised classification performance. While these components lead to effective accuracies and attractive FLOP counts, the added complexity actually makes these transformers slower than their vanilla ViT counterparts. In this paper, we argue that this additional bulk is unnecessary. By pretraining with a strong visual pretext task (MAE), we can strip out all the bells-and-whistles from a state-of-the-art multi-stage vision transformer without losing accuracy. In the process, we create Hiera, an extremely simple hierarchical vision transformer that is more accurate than previous models while being significantly faster both at inference and during training. We evaluate Hiera on a variety of tasks for image and video recognition. Our code and models are available at https://github.com/facebookresearch/hiera.
comment: ICML 2023 Oral version. Code+Models: https://github.com/facebookresearch/hiera
☆ StyleGAN knows Normal, Depth, Albedo, and More
Intrinsic images, in the original sense, are image-like maps of scene properties like depth, normal, albedo or shading. This paper demonstrates that StyleGAN can easily be induced to produce intrinsic images. The procedure is straightforward. We show that, if StyleGAN produces $G({w})$ from latents ${w}$, then for each type of intrinsic image, there is a fixed offset ${d}_c$ so that $G({w}+{d}_c)$ is that type of intrinsic image for $G({w})$. Here ${d}_c$ is {\em independent of ${w}$}. The StyleGAN we used was pretrained by others, so this property is not some accident of our training regime. We show that there are image transformations StyleGAN will {\em not} produce in this fashion, so StyleGAN is not a generic image regression engine. It is conceptually exciting that an image generator should ``know'' and represent intrinsic images. There may also be practical advantages to using a generative model to produce intrinsic images. The intrinsic images obtained from StyleGAN compare well both qualitatively and quantitatively with those obtained by using SOTA image regression techniques; but StyleGAN's intrinsic images are robust to relighting effects, unlike SOTA methods.
comment: Beyond Image Generation: StyleGAN knows Normals, Depth, Albedo, Shading, Segmentation and perhaps more!
☆ Continual Learning for Abdominal Multi-Organ and Tumor Segmentation MICCAI-2023
The ability to dynamically extend a model to new data and classes is critical for multiple organ and tumor segmentation. However, due to privacy regulations, accessing previous data and annotations can be problematic in the medical domain. This poses a significant barrier to preserving the high segmentation accuracy of the old classes when learning from new classes because of the catastrophic forgetting problem. In this paper, we first empirically demonstrate that simply using high-quality pseudo labels can fairly mitigate this problem in the setting of organ segmentation. Furthermore, we put forward an innovative architecture designed specifically for continuous organ and tumor segmentation, which incurs minimal computational overhead. Our proposed design involves replacing the conventional output layer with a suite of lightweight, class-specific heads, thereby offering the flexibility to accommodate newly emerging classes. These heads enable independent predictions for newly introduced and previously learned classes, effectively minimizing the impact of new classes on old ones during the course of continual learning. We further propose incorporating Contrastive Language-Image Pretraining (CLIP) embeddings into the organ-specific heads. These embeddings encapsulate the semantic information of each class, informed by extensive image-text co-training. The proposed method is evaluated on both in-house and public abdominal CT datasets under organ and tumor segmentation tasks. Empirical results suggest that the proposed design improves the segmentation performance of a baseline neural network on newly-introduced and previously-learned classes along the learning trajectory.
comment: MICCAI-2023
☆ Diffusion Self-Guidance for Controllable Image Generation
Large-scale generative models are capable of producing high-quality images from detailed text descriptions. However, many aspects of an image are difficult or impossible to convey through text. We introduce self-guidance, a method that provides greater control over generated images by guiding the internal representations of diffusion models. We demonstrate that properties such as the shape, location, and appearance of objects can be extracted from these representations and used to steer sampling. Self-guidance works similarly to classifier guidance, but uses signals present in the pretrained model itself, requiring no additional models or training. We show how a simple set of properties can be composed to perform challenging image manipulations, such as modifying the position or size of objects, merging the appearance of objects in one image with the layout of another, composing objects from many images into one, and more. We also show that self-guidance can be used to edit real images. For results and an interactive demo, see our project page at https://dave.ml/selfguidance/
comment: Project page at https://dave.ml/selfguidance/
☆ Using generative AI to investigate medical imagery models and datasets
AI models have shown promise in many medical imaging tasks. However, our ability to explain what signals these models have learned is severely lacking. Explanations are needed in order to increase the trust in AI-based models, and could enable novel scientific discovery by uncovering signals in the data that are not yet known to experts. In this paper, we present a method for automatic visual explanations leveraging team-based expertise by generating hypotheses of what visual signals in the images are correlated with the task. We propose the following 4 steps: (i) Train a classifier to perform a given task (ii) Train a classifier guided StyleGAN-based image generator (StylEx) (iii) Automatically detect and visualize the top visual attributes that the classifier is sensitive towards (iv) Formulate hypotheses for the underlying mechanisms, to stimulate future research. Specifically, we present the discovered attributes to an interdisciplinary panel of experts so that hypotheses can account for social and structural determinants of health. We demonstrate results on eight prediction tasks across three medical imaging modalities: retinal fundus photographs, external eye photographs, and chest radiographs. We showcase examples of attributes that capture clinically known features, confounders that arise from factors beyond physiological mechanisms, and reveal a number of physiologically plausible novel attributes. Our approach has the potential to enable researchers to better understand, improve their assessment, and extract new knowledge from AI-based models. Importantly, we highlight that attributes generated by our framework can capture phenomena beyond physiology or pathophysiology, reflecting the real world nature of healthcare delivery and socio-cultural factors. Finally, we intend to release code to enable researchers to train their own StylEx models and analyze their predictive tasks.
comment: 34 pages, 1 figure
☆ SnapFusion: Text-to-Image Diffusion Model on Mobile Devices within Two Seconds
Text-to-image diffusion models can create stunning images from natural language descriptions that rival the work of professional artists and photographers. However, these models are large, with complex network architectures and tens of denoising iterations, making them computationally expensive and slow to run. As a result, high-end GPUs and cloud-based inference are required to run diffusion models at scale. This is costly and has privacy implications, especially when user data is sent to a third party. To overcome these challenges, we present a generic approach that, for the first time, unlocks running text-to-image diffusion models on mobile devices in less than $2$ seconds. We achieve so by introducing efficient network architecture and improving step distillation. Specifically, we propose an efficient UNet by identifying the redundancy of the original model and reducing the computation of the image decoder via data distillation. Further, we enhance the step distillation by exploring training strategies and introducing regularization from classifier-free guidance. Our extensive experiments on MS-COCO show that our model with $8$ denoising steps achieves better FID and CLIP scores than Stable Diffusion v$1.5$ with $50$ steps. Our work democratizes content creation by bringing powerful text-to-image diffusion models to the hands of users.
comment: Our project webpage: https://snap-research.github.io/SnapFusion/
☆ Active Reinforcement Learning under Limited Visual Observability
In this work, we investigate Active Reinforcement Learning (Active-RL), where an embodied agent simultaneously learns action policy for the task while also controlling its visual observations in partially observable environments. We denote the former as motor policy and the latter as sensory policy. For example, humans solve real world tasks by hand manipulation (motor policy) together with eye movements (sensory policy). Active-RL poses challenges on coordinating two policies given their mutual influence. We propose SUGARL, Sensorimotor Understanding Guided Active Reinforcement Learning, a framework that models motor and sensory policies separately, but jointly learns them using with an intrinsic sensorimotor reward. This learnable reward is assigned by sensorimotor reward module, incentivizes the sensory policy to select observations that are optimal to infer its own motor action, inspired by the sensorimotor stage of humans. Through a series of experiments, we show the effectiveness of our method across a range of observability conditions and its adaptability to existed RL algorithms. The sensory policies learned through our method are observed to exhibit effective active vision strategies.
comment: Project page at https://elicassion.github.io/sugarl/sugarl.html Code at https://github.com/elicassion/sugarl Environment library at https://github.com/elicassion/active-gym
☆ Improving and Benchmarking Offline Reinforcement Learning Algorithms
Recently, Offline Reinforcement Learning (RL) has achieved remarkable progress with the emergence of various algorithms and datasets. However, these methods usually focus on algorithmic advancements, ignoring that many low-level implementation choices considerably influence or even drive the final performance. As a result, it becomes hard to attribute the progress in Offline RL as these choices are not sufficiently discussed and aligned in the literature. In addition, papers focusing on a dataset (e.g., D4RL) often ignore algorithms proposed on another dataset (e.g., RL Unplugged), causing isolation among the algorithms, which might slow down the overall progress. Therefore, this work aims to bridge the gaps caused by low-level choices and datasets. To this end, we empirically investigate 20 implementation choices using three representative algorithms (i.e., CQL, CRR, and IQL) and present a guidebook for choosing implementations. Following the guidebook, we find two variants CRR+ and CQL+ , achieving new state-of-the-art on D4RL. Moreover, we benchmark eight popular offline RL algorithms across datasets under unified training and evaluation framework. The findings are inspiring: the success of a learning paradigm severely depends on the data distribution, and some previous conclusions are biased by the dataset used. Our code is available at https://github.com/sail-sg/offbench.
comment: preprint
☆ Dynamic Algorithms for Matroid Submodular Maximization
Submodular maximization under matroid and cardinality constraints are classical problems with a wide range of applications in machine learning, auction theory, and combinatorial optimization. In this paper, we consider these problems in the dynamic setting where (1) we have oracle access to a monotone submodular function $f: 2^{V} \rightarrow \mathbb{R}^+$ and (2) we are given a sequence $\mathcal{S}$ of insertions and deletions of elements of an underlying ground set $V$. We develop the first parameterized (by the rank $k$ of a matroid $\mathcal{M}$) dynamic $(4+\epsilon)$-approximation algorithm for the submodular maximization problem under the matroid constraint using an expected worst-case $O(k\log(k)\log^3{(k/\epsilon)})$ query complexity where $0 < \epsilon \le 1$. Chen and Peng at STOC'22 studied the complexity of this problem in the insertion-only dynamic model (a restricted version of the fully dynamic model where deletion is not allowed), and they raised the following important open question: *"for fully dynamic streams [sequences of insertions and deletions of elements], there is no known constant-factor approximation algorithm with poly(k) amortized queries for matroid constraints."* Our dynamic algorithm answers this question as well as an open problem of Lattanzi et al. (NeurIPS'20) affirmatively. As a byproduct, for the submodular maximization under the cardinality constraint $k$, we propose a parameterized (by the cardinality constraint $k$) dynamic algorithm that maintains a $(2+\epsilon)$-approximate solution of the sequence $\mathcal{S}$ at any time $t$ using the expected amortized worst-case complexity $O(k\epsilon^{-1}\log^2(k))$. This is the first dynamic algorithm for the problem that has a query complexity independent of the size of ground set $V$.
comment: Abstract shortened due to arxiv's requirements
☆ LIV: Language-Image Representations and Rewards for Robotic Control ICML 2023
We present Language-Image Value learning (LIV), a unified objective for vision-language representation and reward learning from action-free videos with text annotations. Exploiting a novel connection between dual reinforcement learning and mutual information contrastive learning, the LIV objective trains a multi-modal representation that implicitly encodes a universal value function for tasks specified as language or image goals. We use LIV to pre-train the first control-centric vision-language representation from large human video datasets such as EpicKitchen. Given only a language or image goal, the pre-trained LIV model can assign dense rewards to each frame in videos of unseen robots or humans attempting that task in unseen environments. Further, when some target domain-specific data is available, the same objective can be used to fine-tune and improve LIV and even other pre-trained representations for robotic control and reward specification in that domain. In our experiments on several simulated and real-world robot environments, LIV models consistently outperform the best prior input state representations for imitation learning, as well as reward specification methods for policy synthesis. Our results validate the advantages of joint vision-language representation and reward learning within the unified, compact LIV framework.
comment: Extended version of ICML 2023 camera-ready; Project website: https://penn-pal-lab.github.io/LIV/
☆ Speaker-specific Thresholding for Robust Imposter Identification in Unseen Speaker Recognition
Speaker identification systems are deployed in diverse environments, often different from the lab conditions on which they are trained and tested. In this paper, first, we show the problem of generalization using fixed thresholds computed using the equal error rate metric. Secondly, we introduce a novel and generalizable speaker-specific thresholding technique for robust imposter identification in unseen speaker identification. We propose a speaker-specific adaptive threshold, which can be computed using the enrollment audio samples, for identifying imposters in unseen speaker identification. Furthermore, we show the efficacy of the proposed technique on VoxCeleb1, VCTK and the FFSVC 2022 datasets, beating the baseline fixed thresholding by up to 25%. Finally, we exhibit that the proposed algorithm is also generalizable, demonstrating its performance on ResNet50, ECAPA-TDNN and RawNet3 speaker encoders.
☆ Differential Diffusion: Giving Each Pixel Its Strength
Text-based image editing has advanced significantly in recent years. With the rise of diffusion models, image editing via textual instructions has become ubiquitous. Unfortunately, current models lack the ability to customize the quantity of the change per pixel or per image fragment, resorting to changing the entire image in an equal amount, or editing a specific region using a binary mask. In this paper, we suggest a new framework which enables the user to customize the quantity of change for each image fragment, thereby enhancing the flexibility and verbosity of modern diffusion models. Our framework does not require model training or fine-tuning, but instead performs everything at inference time, making it easily applicable to an existing model. We show both qualitatively and quantitatively that our method allows better controllability and can produce results which are unattainable by existing models. Our code is available at: https://github.com/exx8/differential-diffusion
comment: Our code is available at: https://github.com/exx8/differential-diffusion
☆ EEL: Efficiently Encoding Lattices for Reranking ACL 2023
Standard decoding approaches for conditional text generation tasks typically search for an output hypothesis with high model probability, but this may not yield the best hypothesis according to human judgments of quality. Reranking to optimize for "downstream" metrics can better optimize for quality, but many metrics of interest are computed with pre-trained language models, which are slow to apply to large numbers of hypotheses. We explore an approach for reranking hypotheses by using Transformers to efficiently encode lattices of generated outputs, a method we call EEL. With a single Transformer pass over the entire lattice, we can approximately compute a contextualized representation of each token as if it were only part of a single hypothesis in isolation. We combine this approach with a new class of token-factored rerankers (TFRs) that allow for efficient extraction of high reranker-scoring hypotheses from the lattice. Empirically, our approach incurs minimal degradation error compared to the exponentially slower approach of encoding each hypothesis individually. When applying EEL with TFRs across three text generation tasks, our results show both substantial speedup compared to naive reranking and often better performance on downstream metrics than comparable approaches.
comment: ACL 2023 (16 pages), code available at https://github.com/PrasannS/eel-reranking
☆ Exposing Attention Glitches with Flip-Flop Language Modeling
Why do large language models sometimes output factual inaccuracies and exhibit erroneous reasoning? The brittleness of these models, particularly when executing long chains of reasoning, currently seems to be an inevitable price to pay for their advanced capabilities of coherently synthesizing knowledge, pragmatics, and abstract thought. Towards making sense of this fundamentally unsolved problem, this work identifies and analyzes the phenomenon of attention glitches, in which the Transformer architecture's inductive biases intermittently fail to capture robust reasoning. To isolate the issue, we introduce flip-flop language modeling (FFLM), a parametric family of synthetic benchmarks designed to probe the extrapolative behavior of neural language models. This simple generative task requires a model to copy binary symbols over long-range dependencies, ignoring the tokens in between. We find that Transformer FFLMs suffer from a long tail of sporadic reasoning errors, some of which we can eliminate using various regularization techniques. Our preliminary mechanistic analyses show why the remaining errors may be very difficult to diagnose and resolve. We hypothesize that attention glitches account for (some of) the closed-domain hallucinations in natural LLMs.
☆ CS4ML: A general framework for active learning with arbitrary data based on Christoffel functions
We introduce a general framework for active learning in regression problems. Our framework extends the standard setup by allowing for general types of data, rather than merely pointwise samples of the target function. This generalization covers many cases of practical interest, such as data acquired in transform domains (e.g., Fourier data), vector-valued data (e.g., gradient-augmented data), data acquired along continuous curves, and, multimodal data (i.e., combinations of different types of measurements). Our framework considers random sampling according to a finite number of sampling measures and arbitrary nonlinear approximation spaces (model classes). We introduce the concept of generalized Christoffel functions and show how these can be used to optimize the sampling measures. We prove that this leads to near-optimal sample complexity in various important cases. This paper focuses on applications in scientific computing, where active learning is often desirable, since it is usually expensive to generate data. We demonstrate the efficacy of our framework for gradient-augmented learning with polynomials, Magnetic Resonance Imaging (MRI) using generative models and adaptive sampling for solving PDEs using Physics-Informed Neural Networks (PINNs).
☆ Train Offline, Test Online: A Real Robot Learning Benchmark ICRA 2023
Three challenges limit the progress of robot learning research: robots are expensive (few labs can participate), everyone uses different robots (findings do not generalize across labs), and we lack internet-scale robotics data. We take on these challenges via a new benchmark: Train Offline, Test Online (TOTO). TOTO provides remote users with access to shared robotic hardware for evaluating methods on common tasks and an open-source dataset of these tasks for offline training. Its manipulation task suite requires challenging generalization to unseen objects, positions, and lighting. We present initial results on TOTO comparing five pretrained visual representations and four offline policy learning baselines, remotely contributed by five institutions. The real promise of TOTO, however, lies in the future: we release the benchmark for additional submissions from any user, enabling easy, direct comparison to several methods without the need to obtain hardware or collect data.
comment: Accepted to ICRA 2023
☆ STEVE-1: A Generative Model for Text-to-Behavior in Minecraft
Constructing AI models that respond to text instructions is challenging, especially for sequential decision-making tasks. This work introduces an instruction-tuned Video Pretraining (VPT) model for Minecraft called STEVE-1, demonstrating that the unCLIP approach, utilized in DALL-E 2, is also effective for creating instruction-following sequential decision-making agents. STEVE-1 is trained in two steps: adapting the pretrained VPT model to follow commands in MineCLIP's latent space, then training a prior to predict latent codes from text. This allows us to finetune VPT through self-supervised behavioral cloning and hindsight relabeling, bypassing the need for costly human text annotations. By leveraging pretrained models like VPT and MineCLIP and employing best practices from text-conditioned image generation, STEVE-1 costs just $60 to train and can follow a wide range of short-horizon open-ended text and visual instructions in Minecraft. STEVE-1 sets a new bar for open-ended instruction following in Minecraft with low-level controls (mouse and keyboard) and raw pixel inputs, far outperforming previous baselines. We provide experimental evidence highlighting key factors for downstream performance, including pretraining, classifier-free guidance, and data scaling. All resources, including our model weights, training scripts, and evaluation tools are made available for further research.
☆ Interpreting GNN-based IDS Detections Using Provenance Graph Structural Features
The black-box nature of complex Neural Network (NN)-based models has hindered their widespread adoption in security domains due to the lack of logical explanations and actionable follow-ups for their predictions. To enhance the transparency and accountability of Graph Neural Network (GNN) security models used in system provenance analysis, we propose PROVEXPLAINER, a framework for projecting abstract GNN decision boundaries onto interpretable feature spaces. We first replicate the decision-making process of GNNbased security models using simpler and explainable models such as Decision Trees (DTs). To maximize the accuracy and fidelity of the surrogate models, we propose novel graph structural features founded on classical graph theory and enhanced by extensive data study with security domain knowledge. Our graph structural features are closely tied to problem-space actions in the system provenance domain, which allows the detection results to be explained in descriptive, human language. PROVEXPLAINER allowed simple DT models to achieve 95% fidelity to the GNN on program classification tasks with general graph structural features, and 99% fidelity on malware detection tasks with a task-specific feature package tailored for direct interpretation. The explanations for malware classification are demonstrated with case studies of five real-world malware samples across three malware families.
☆ Second Sight: Using brain-optimized encoding models to align image distributions with human brain activity
Two recent developments have accelerated progress in image reconstruction from human brain activity: large datasets that offer samples of brain activity in response to many thousands of natural scenes, and the open-sourcing of powerful stochastic image-generators that accept both low- and high-level guidance. Most work in this space has focused on obtaining point estimates of the target image, with the ultimate goal of approximating literal pixel-wise reconstructions of target images from the brain activity patterns they evoke. This emphasis belies the fact that there is always a family of images that are equally compatible with any evoked brain activity pattern, and the fact that many image-generators are inherently stochastic and do not by themselves offer a method for selecting the single best reconstruction from among the samples they generate. We introduce a novel reconstruction procedure (Second Sight) that iteratively refines an image distribution to explicitly maximize the alignment between the predictions of a voxel-wise encoding model and the brain activity patterns evoked by any target image. We show that our process converges on a distribution of high-quality reconstructions by refining both semantic content and low-level image details across iterations. Images sampled from these converged image distributions are competitive with state-of-the-art reconstruction algorithms. Interestingly, the time-to-convergence varies systematically across visual cortex, with earlier visual areas generally taking longer and converging on narrower image distributions, relative to higher-level brain areas. Second Sight thus offers a succinct and novel method for exploring the diversity of representations across visual brain areas.
comment: 15 Figures, 19 pages including the appendix
☆ Minding Language Models' (Lack of) Theory of Mind: A Plug-and-Play Multi-Character Belief Tracker
Theory of Mind (ToM)$\unicode{x2014}$the ability to reason about the mental states of other people$\unicode{x2014}$is a key element of our social intelligence. Yet, despite their ever more impressive performance, large-scale neural language models still lack basic theory of mind capabilities out-of-the-box. We posit that simply scaling up models will not imbue them with theory of mind due to the inherently symbolic and implicit nature of the phenomenon, and instead investigate an alternative: can we design a decoding-time algorithm that enhances theory of mind of off-the-shelf neural language models without explicit supervision? We present SymbolicToM, a plug-and-play approach to reason about the belief states of multiple characters in reading comprehension tasks via explicit symbolic representation. More concretely, our approach tracks each entity's beliefs, their estimation of other entities' beliefs, and higher-order levels of reasoning, all through graphical representations, allowing for more precise and interpretable reasoning than previous approaches. Empirical results on the well-known ToMi benchmark (Le et al., 2019) demonstrate that SymbolicToM dramatically enhances off-the-shelf neural networks' theory of mind in a zero-shot setting while showing robust out-of-distribution performance compared to supervised baselines. Our work also reveals spurious patterns in existing theory of mind benchmarks, emphasizing the importance of out-of-distribution evaluation and methods that do not overfit a particular dataset.
☆ Better Private Linear Regression Through Better Private Feature Selection
Existing work on differentially private linear regression typically assumes that end users can precisely set data bounds or algorithmic hyperparameters. End users often struggle to meet these requirements without directly examining the data (and violating privacy). Recent work has attempted to develop solutions that shift these burdens from users to algorithms, but they struggle to provide utility as the feature dimension grows. This work extends these algorithms to higher-dimensional problems by introducing a differentially private feature selection method based on Kendall rank correlation. We prove a utility guarantee for the setting where features are normally distributed and conduct experiments across 25 datasets. We find that adding this private feature selection step before regression significantly broadens the applicability of ``plug-and-play'' private linear regression algorithms at little additional cost to privacy, computation, or decision-making by the end user.
☆ The feasibility of artificial consciousness through the lens of neuroscience
Interactions with large language models have led to the suggestion that these models may be conscious. From the perspective of neuroscience, this position is difficult to defend. For one, the architecture of large language models is missing key features of the thalamocortical system that have been linked to conscious awareness in mammals. Secondly, the inputs to large language models lack the embodied, embedded information content characteristic of our sensory contact with the world around us. Finally, while the previous two arguments can be overcome in future AI systems, the third one might be harder to bridge in the near future. Namely, we argue that consciousness might depend on having 'skin in the game', in that the existence of the system depends on its actions, which is not true for present-day artificial intelligence.
☆ Interaction Measures, Partition Lattices and Kernel Tests for High-Order Interactions
Models that rely solely on pairwise relationships often fail to capture the complete statistical structure of the complex multivariate data found in diverse domains, such as socio-economic, ecological, or biomedical systems. Non-trivial dependencies between groups of more than two variables can play a significant role in the analysis and modelling of such systems, yet extracting such high-order interactions from data remains challenging. Here, we introduce a hierarchy of $d$-order ($d \geq 2$) interaction measures, increasingly inclusive of possible factorisations of the joint probability distribution, and define non-parametric, kernel-based tests to establish systematically the statistical significance of $d$-order interactions. We also establish mathematical links with lattice theory, which elucidate the derivation of the interaction measures and their composite permutation tests; clarify the connection of simplicial complexes with kernel matrix centring; and provide a means to enhance computational efficiency. We illustrate our results numerically with validations on synthetic data, and through an application to neuroimaging data.
comment: 19 pages, 7 figures
☆ SpotTarget: Rethinking the Effect of Target Edges for Link Prediction in Graph Neural Networks
Graph Neural Networks (GNNs) have demonstrated promising outcomes across various tasks, including node classification and link prediction. Despite their remarkable success in various high-impact applications, we have identified three common pitfalls in message passing for link prediction. Particularly, in prevalent GNN frameworks (e.g., DGL and PyTorch-Geometric), the target edges (i.e., the edges being predicted) consistently exist as message passing edges in the graph during training. Consequently, this results in overfitting and distribution shift, both of which adversely impact the generalizability to test the target edges. Additionally, during test time, the failure to exclude the test target edges leads to implicit test leakage caused by neighborhood aggregation. In this paper, we analyze these three pitfalls and investigate the impact of including or excluding target edges on the performance of nodes with varying degrees during training and test phases. Our theoretical and empirical analysis demonstrates that low-degree nodes are more susceptible to these pitfalls. These pitfalls can have detrimental consequences when GNNs are implemented in production systems. To systematically address these pitfalls, we propose SpotTarget, an effective and efficient GNN training framework. During training, SpotTarget leverages our insight regarding low-degree nodes and excludes train target edges connected to at least one low-degree node. During test time, it emulates real-world scenarios of GNN usage in production and excludes all test target edges. Our experiments conducted on diverse real-world datasets, demonstrate that SpotTarget significantly enhances GNNs, achieving up to a 15x increase in accuracy in sparse graphs. Furthermore, SpotTarget consistently and dramatically improves the performance for low-degree nodes in dense graphs.
comment: 8 pages, 2 page appendix
☆ Domain Generalization for Domain-Linked Classes
Domain generalization (DG) focuses on transferring domain-invariant knowledge from multiple source domains (available at train time) to an, a priori, unseen target domain(s). This requires a class to be expressed in multiple domains for the learning algorithm to break the spurious correlations between domain and class. However, in the real-world, classes may often be domain-linked, i.e. expressed only in a specific domain, which leads to extremely poor generalization performance for these classes. In this work, we aim to learn generalizable representations for these domain-linked classes by transferring domain-invariant knowledge from classes expressed in multiple source domains (domain-shared classes). To this end, we introduce this task to the community and propose a Fair and cONtrastive feature-space regularization algorithm for Domain-linked DG, FOND. Rigorous and reproducible experiments with baselines across popular DG tasks demonstrate our method and its variants' ability to accomplish state-of-the-art DG results for domain-linked classes. We also provide practical insights on data conditions that increase domain-linked class generalizability to tackle real-world data scarcity.
☆ Quantifying Deep Learning Model Uncertainty in Conformal Prediction
Precise estimation of predictive uncertainty in deep neural networks is a critical requirement for reliable decision-making in machine learning and statistical modeling, particularly in the context of medical AI. Conformal Prediction (CP) has emerged as a promising framework for representing the model uncertainty by providing well-calibrated confidence levels for individual predictions. However, the quantification of model uncertainty in conformal prediction remains an active research area, yet to be fully addressed. In this paper, we explore state-of-the-art CP methodologies and their theoretical foundations. We propose a probabilistic approach in quantifying the model uncertainty derived from the produced prediction sets in conformal prediction and provide certified boundaries for the computed uncertainty. By doing so, we allow model uncertainty measured by CP to be compared by other uncertainty quantification methods such as Bayesian (e.g., MC-Dropout and DeepEnsemble) and Evidential approaches.
comment: 7 pages, 4 figures
☆ Is novelty predictable?
Machine learning-based design has gained traction in the sciences, most notably in the design of small molecules, materials, and proteins, with societal implications spanning drug development and manufacturing, plastic degradation, and carbon sequestration. When designing objects to achieve novel property values with machine learning, one faces a fundamental challenge: how to push past the frontier of current knowledge, distilled from the training data into the model, in a manner that rationally controls the risk of failure. If one trusts learned models too much in extrapolation, one is likely to design rubbish. In contrast, if one does not extrapolate, one cannot find novelty. Herein, we ponder how one might strike a useful balance between these two extremes. We focus in particular on designing proteins with novel property values, although much of our discussion addresses machine learning-based design more broadly.
☆ IQL-TD-MPC: Implicit Q-Learning for Hierarchical Model Predictive Control
Model-based reinforcement learning (RL) has shown great promise due to its sample efficiency, but still struggles with long-horizon sparse-reward tasks, especially in offline settings where the agent learns from a fixed dataset. We hypothesize that model-based RL agents struggle in these environments due to a lack of long-term planning capabilities, and that planning in a temporally abstract model of the environment can alleviate this issue. In this paper, we make two key contributions: 1) we introduce an offline model-based RL algorithm, IQL-TD-MPC, that extends the state-of-the-art Temporal Difference Learning for Model Predictive Control (TD-MPC) with Implicit Q-Learning (IQL); 2) we propose to use IQL-TD-MPC as a Manager in a hierarchical setting with any off-the-shelf offline RL algorithm as a Worker. More specifically, we pre-train a temporally abstract IQL-TD-MPC Manager to predict "intent embeddings", which roughly correspond to subgoals, via planning. We empirically show that augmenting state representations with intent embeddings generated by an IQL-TD-MPC manager significantly improves off-the-shelf offline RL agents' performance on some of the most challenging D4RL benchmark tasks. For instance, the offline RL algorithms AWAC, TD3-BC, DT, and CQL all get zero or near-zero normalized evaluation scores on the medium and large antmaze tasks, while our modification gives an average score over 40.
☆ A Transformer-based representation-learning model with unified processing of multimodal input for clinical diagnostics
During the diagnostic process, clinicians leverage multimodal information, such as chief complaints, medical images, and laboratory-test results. Deep-learning models for aiding diagnosis have yet to meet this requirement. Here we report a Transformer-based representation-learning model as a clinical diagnostic aid that processes multimodal input in a unified manner. Rather than learning modality-specific features, the model uses embedding layers to convert images and unstructured and structured text into visual tokens and text tokens, and bidirectional blocks with intramodal and intermodal attention to learn a holistic representation of radiographs, the unstructured chief complaint and clinical history, structured clinical information such as laboratory-test results and patient demographic information. The unified model outperformed an image-only model and non-unified multimodal diagnosis models in the identification of pulmonary diseases (by 12% and 9%, respectively) and in the prediction of adverse clinical outcomes in patients with COVID-19 (by 29% and 7%, respectively). Leveraging unified multimodal Transformer-based models may help streamline triage of patients and facilitate the clinical decision process.
comment: Accepted by Nature Biomedical Engineering
☆ Non-stationary Reinforcement Learning under General Function Approximation ICML 2023
General function approximation is a powerful tool to handle large state and action spaces in a broad range of reinforcement learning (RL) scenarios. However, theoretical understanding of non-stationary MDPs with general function approximation is still limited. In this paper, we make the first such an attempt. We first propose a new complexity metric called dynamic Bellman Eluder (DBE) dimension for non-stationary MDPs, which subsumes majority of existing tractable RL problems in static MDPs as well as non-stationary MDPs. Based on the proposed complexity metric, we propose a novel confidence-set based model-free algorithm called SW-OPEA, which features a sliding window mechanism and a new confidence set design for non-stationary MDPs. We then establish an upper bound on the dynamic regret for the proposed algorithm, and show that SW-OPEA is provably efficient as long as the variation budget is not significantly large. We further demonstrate via examples of non-stationary linear and tabular MDPs that our algorithm performs better in small variation budget scenario than the existing UCB-type algorithms. To the best of our knowledge, this is the first dynamic regret analysis in non-stationary MDPs with general function approximation.
comment: ICML 2023
☆ Loss-Optimal Classification Trees: A Generalized Framework and the Logistic Case
The Classification Tree (CT) is one of the most common models in interpretable machine learning. Although such models are usually built with greedy strategies, in recent years, thanks to remarkable advances in Mixer-Integer Programming (MIP) solvers, several exact formulations of the learning problem have been developed. In this paper, we argue that some of the most relevant ones among these training models can be encapsulated within a general framework, whose instances are shaped by the specification of loss functions and regularizers. Next, we introduce a novel realization of this framework: specifically, we consider the logistic loss, handled in the MIP setting by a linear piece-wise approximation, and couple it with $\ell_1$-regularization terms. The resulting Optimal Logistic Tree model numerically proves to be able to induce trees with enhanced interpretability features and competitive generalization capabilities, compared to the state-of-the-art MIP-based approaches.
☆ A deep-learning approach to early identification of suggested sexual harassment from videos
Sexual harassment, sexual abuse, and sexual violence are prevalent problems in this day and age. Women's safety is an important issue that needs to be highlighted and addressed. Given this issue, we have studied each of these concerns and the factors that affect it based on images generated from movies. We have classified the three terms (harassment, abuse, and violence) based on the visual attributes present in images depicting these situations. We identified that factors such as facial expression of the victim and perpetrator and unwanted touching had a direct link to identifying the scenes containing sexual harassment, abuse and violence. We also studied and outlined how state-of-the-art explicit content detectors such as Google Cloud Vision API and Clarifai API fail to identify and categorise these images. Based on these definitions and characteristics, we have developed a first-of-its-kind dataset from various Indian movie scenes. These scenes are classified as sexual harassment, sexual abuse, or sexual violence and exported in the PASCAL VOC 1.1 format. Our dataset is annotated on the identified relevant features and can be used to develop and train a deep-learning computer vision model to identify these issues. The dataset is publicly available for research and development.
comment: 7 pages, 4 figures
☆ BitE : Accelerating Learned Query Optimization in a Mixed-Workload Environment
Although the many efforts to apply deep reinforcement learning to query optimization in recent years, there remains room for improvement as query optimizers are complex entities that require hand-designed tuning of workloads and datasets. Recent research present learned query optimizations results mostly in bulks of single workloads which focus on picking up the unique traits of the specific workload. This proves to be problematic in scenarios where the different characteristics of multiple workloads and datasets are to be mixed and learned together. Henceforth, in this paper, we propose BitE, a novel ensemble learning model using database statistics and metadata to tune a learned query optimizer for enhancing performance. On the way, we introduce multiple revisions to solve several challenges: we extend the search space for the optimal Abstract SQL Plan(represented as a JSON object called ASP) by expanding hintsets, we steer the model away from the default plans that may be biased by configuring the experience with all unique plans of queries, and we deviate from the traditional loss functions and choose an alternative method to cope with underestimation and overestimation of reward. Our model achieves 19.6% more improved queries and 15.8% less regressed queries compared to the existing traditional methods whilst using a comparable level of resources.
comment: This work was done when the first three author were interns in SAP Labs Korea and they have equal contribution
☆ What model does MuZero learn?
Model-based reinforcement learning has drawn considerable interest in recent years, given its promise to improve sample efficiency. Moreover, when using deep-learned models, it is potentially possible to learn compact models from complex sensor data. However, the effectiveness of these learned models, particularly their capacity to plan, i.e., to improve the current policy, remains unclear. In this work, we study MuZero, a well-known deep model-based reinforcement learning algorithm, and explore how far it achieves its learning objective of a value-equivalent model and how useful the learned models are for policy improvement. Amongst various other insights, we conclude that the model learned by MuZero cannot effectively generalize to evaluate unseen policies, which limits the extent to which we can additionally improve the current policy by planning with the model.
☆ Deformable Convolutions and LSTM-based Flexible Event Frame Fusion Network for Motion Deblurring
Event cameras differ from conventional RGB cameras in that they produce asynchronous data sequences. While RGB cameras capture every frame at a fixed rate, event cameras only capture changes in the scene, resulting in sparse and asynchronous data output. Despite the fact that event data carries useful information that can be utilized in motion deblurring of RGB cameras, integrating event and image information remains a challenge. Recent state-of-the-art CNN-based deblurring solutions produce multiple 2-D event frames based on the accumulation of event data over a time period. In most of these techniques, however, the number of event frames is fixed and predefined, which reduces temporal resolution drastically, particularly for scenarios when fast-moving objects are present or when longer exposure times are required. It is also important to note that recent modern cameras (e.g., cameras in mobile phones) dynamically set the exposure time of the image, which presents an additional problem for networks developed for a fixed number of event frames. A Long Short-Term Memory (LSTM)-based event feature extraction module has been developed for addressing these challenges, which enables us to use a dynamically varying number of event frames. Using these modules, we constructed a state-of-the-art deblurring network, Deformable Convolutions and LSTM-based Flexible Event Frame Fusion Network (DLEFNet). It is particularly useful for scenarios in which exposure times vary depending on factors such as lighting conditions or the presence of fast-moving objects in the scene. It has been demonstrated through evaluation results that the proposed method can outperform the existing state-of-the-art networks for deblurring task in synthetic and real-world data sets.
☆ When Does Bottom-up Beat Top-down in Hierarchical Community Detection?
Hierarchical clustering of networks consists in finding a tree of communities, such that lower levels of the hierarchy reveal finer-grained community structures. There are two main classes of algorithms tackling this problem. Divisive ($\textit{top-down}$) algorithms recursively partition the nodes into two communities, until a stopping rule indicates that no further split is needed. In contrast, agglomerative ($\textit{bottom-up}$) algorithms first identify the smallest community structure and then repeatedly merge the communities using a $\textit{linkage}$ method. In this article, we establish theoretical guarantees for the recovery of the hierarchical tree and community structure of a Hierarchical Stochastic Block Model by a bottom-up algorithm. We also establish that this bottom-up algorithm attains the information-theoretic threshold for exact recovery at intermediate levels of the hierarchy. Notably, these recovery conditions are less restrictive compared to those existing for top-down algorithms. This shows that bottom-up algorithms extend the feasible region for achieving exact recovery at intermediate levels. Numerical experiments on both synthetic and real data sets confirm the superiority of bottom-up algorithms over top-down algorithms. We also observe that top-down algorithms can produce dendrograms with inversions. These findings contribute to a better understanding of hierarchical clustering techniques and their applications in network analysis.
☆ In or Out? Fixing ImageNet Out-of-Distribution Detection Evaluation ICML 2023
Out-of-distribution (OOD) detection is the problem of identifying inputs which are unrelated to the in-distribution task. The OOD detection performance when the in-distribution (ID) is ImageNet-1K is commonly being tested on a small range of test OOD datasets. We find that most of the currently used test OOD datasets, including datasets from the open set recognition (OSR) literature, have severe issues: In some cases more than 50$\%$ of the dataset contains objects belonging to one of the ID classes. These erroneous samples heavily distort the evaluation of OOD detectors. As a solution, we introduce with NINCO a novel test OOD dataset, each sample checked to be ID free, which with its fine-grained range of OOD classes allows for a detailed analysis of an OOD detector's strengths and failure modes, particularly when paired with a number of synthetic "OOD unit-tests". We provide detailed evaluations across a large set of architectures and OOD detection methods on NINCO and the unit-tests, revealing new insights about model weaknesses and the effects of pretraining on OOD detection performance. We provide code and data at https://github.com/j-cb/NINCO.
comment: ICML 2023. Datasets, code and evaluation data at https://github.com/j-cb/NINCO
☆ Vocos: Closing the gap between time-domain and Fourier-based neural vocoders for high-quality audio synthesis
Recent advancements in neural vocoding are predominantly driven by Generative Adversarial Networks (GANs) operating in the time-domain. While effective, this approach neglects the inductive bias offered by time-frequency representations, resulting in reduntant and computionally-intensive upsampling operations. Fourier-based time-frequency representation is an appealing alternative, aligning more accurately with human auditory perception, and benefitting from well-established fast algorithms for its computation. Nevertheless, direct reconstruction of complex-valued spectrograms has been historically problematic, primarily due to phase recovery issues. This study seeks to close this gap by presenting Vocos, a new model that addresses the key challenges of modeling spectral coefficients. Vocos demonstrates improved computational efficiency, achieving an order of magnitude increase in speed compared to prevailing time-domain neural vocoding approaches. As shown by objective evaluation, Vocos not only matches state-of-the-art audio quality, but thanks to frequency-aware generator, also effectively mitigates the periodicity issues frequently associated with time-domain GANs. The source code and model weights have been open-sourced at https://github.com/charactr-platform/vocos.
☆ Deep Operator Learning-based Surrogate Models with Uncertainty Quantification for Optimizing Internal Cooling Channel Rib Profiles
This paper designs surrogate models with uncertainty quantification capabilities to improve the thermal performance of rib-turbulated internal cooling channels effectively. To construct the surrogate, we use the deep operator network (DeepONet) framework, a novel class of neural networks designed to approximate mappings between infinite-dimensional spaces using relatively small datasets. The proposed DeepONet takes an arbitrary continuous rib geometry with control points as input and outputs continuous detailed information about the distribution of pressure and heat transfer around the profiled ribs. The datasets needed to train and test the proposed DeepONet framework were obtained by simulating a 2D rib-roughened internal cooling channel. To accomplish this, we continuously modified the input rib geometry by adjusting the control points according to a simple random distribution with constraints, rather than following a predefined path or sampling method. The studied channel has a hydraulic diameter, Dh, of 66.7 mm, and a length-to-hydraulic diameter ratio, L/Dh, of 10. The ratio of rib center height to hydraulic diameter (e/Dh), which was not changed during the rib profile update, was maintained at a constant value of 0.048. The ribs were placed in the channel with a pitch-to-height ratio (P/e) of 10. In addition, we provide the proposed surrogates with effective uncertainty quantification capabilities. This is achieved by converting the DeepONet framework into a Bayesian DeepONet (B-DeepONet). B-DeepONet samples from the posterior distribution of DeepONet parameters using the novel framework of stochastic gradient replica-exchange MCMC.
comment: 25 pages, 12 figures, 4 tables- submitted to the International Journal of Heat and Mass Transfer
☆ Initial Guessing Bias: How Untrained Networks Favor Some Classes
The initial state of neural networks plays a central role in conditioning the subsequent training dynamics. In the context of classification problems, we provide a theoretical analysis demonstrating that the structure of a neural network can condition the model to assign all predictions to the same class, even before the beginning of training, and in the absence of explicit biases. We show that the presence of this phenomenon, which we call "Initial Guessing Bias" (IGB), depends on architectural choices such as activation functions, max-pooling layers, and network depth. Our analysis of IGB has practical consequences, in that it guides architecture selection and initialization. We also highlight theoretical consequences, such as the breakdown of node-permutation symmetry, the violation of self-averaging, the validity of some mean-field approximations, and the non-trivial differences arising with depth.
☆ Birth of a Transformer: A Memory Viewpoint
Large language models based on transformers have achieved great empirical successes. However, as they are deployed more widely, there is a growing need to better understand their internal mechanisms in order to make them more reliable. These models appear to store vast amounts of knowledge from their training data, and to adapt quickly to new information provided in their context or prompt. We study how transformers balance these two types of knowledge by considering a synthetic setup where tokens are generated from either global or context-specific bigram distributions. By a careful empirical analysis of the training process on a simplified two-layer transformer, we illustrate the fast learning of global bigrams and the slower development of an "induction head" mechanism for the in-context bigrams. We highlight the role of weight matrices as associative memories, provide theoretical insights on how gradients enable their learning during training, and study the role of data-distributional properties.
☆ SlothSpeech: Denial-of-service Attack Against Speech Recognition Models
Deep Learning (DL) models have been popular nowadays to execute different speech-related tasks, including automatic speech recognition (ASR). As ASR is being used in different real-time scenarios, it is important that the ASR model remains efficient against minor perturbations to the input. Hence, evaluating efficiency robustness of the ASR model is the need of the hour. We show that popular ASR models like Speech2Text model and Whisper model have dynamic computation based on different inputs, causing dynamic efficiency. In this work, we propose SlothSpeech, a denial-of-service attack against ASR models, which exploits the dynamic behaviour of the model. SlothSpeech uses the probability distribution of the output text tokens to generate perturbations to the audio such that efficiency of the ASR model is decreased. We find that SlothSpeech generated inputs can increase the latency up to 40X times the latency induced by benign input.
☆ Understanding Augmentation-based Self-Supervised Representation Learning via RKHS Approximation
Good data augmentation is one of the key factors that lead to the empirical success of self-supervised representation learning such as contrastive learning and masked language modeling, yet theoretical understanding of its role in learning good representations remains limited. Recent work has built the connection between self-supervised learning and approximating the top eigenspace of a graph Laplacian operator. Learning a linear probe on top of such features can naturally be connected to RKHS regression. In this work, we use this insight to perform a statistical analysis of augmentation-based pretraining. We start from the isometry property, a key geometric characterization of the target function given by the augmentation. Our first main theorem provides, for an arbitrary encoder, near tight bounds for both the estimation error incurred by fitting the linear probe on top of the encoder, and the approximation error entailed by the fitness of the RKHS the encoder learns. Our second main theorem specifically addresses the case where the encoder extracts the top-d eigenspace of a Monte-Carlo approximation of the underlying kernel with the finite pretraining samples. Our analysis completely disentangles the effects of the model and the augmentation. A key ingredient in our analysis is the augmentation complexity, which we use to quantitatively compare different augmentations and analyze their impact on downstream performance on synthetic and real datasets.
comment: 33 pages
☆ Data Interpolants -- That's What Discriminators in Higher-order Gradient-regularized GANs Are
We consider the problem of optimizing the discriminator in generative adversarial networks (GANs) subject to higher-order gradient regularization. We show analytically, via the least-squares (LSGAN) and Wasserstein (WGAN) GAN variants, that the discriminator optimization problem is one of interpolation in $n$-dimensions. The optimal discriminator, derived using variational Calculus, turns out to be the solution to a partial differential equation involving the iterated Laplacian or the polyharmonic operator. The solution is implementable in closed-form via polyharmonic radial basis function (RBF) interpolation. In view of the polyharmonic connection, we refer to the corresponding GANs as Poly-LSGAN and Poly-WGAN. Through experimental validation on multivariate Gaussians, we show that implementing the optimal RBF discriminator in closed-form, with penalty orders $m \approx\lceil \frac{n}{2} \rceil $, results in superior performance, compared to training GAN with arbitrarily chosen discriminator architectures. We employ the Poly-WGAN discriminator to model the latent space distribution of the data with encoder-decoder-based GAN flavors such as Wasserstein autoencoders.
☆ An End-to-End Time Series Model for Simultaneous Imputation and Forecast
Time series forecasting using historical data has been an interesting and challenging topic, especially when the data is corrupted by missing values. In many industrial problem, it is important to learn the inference function between the auxiliary observations and target variables as it provides additional knowledge when the data is not fully observed. We develop an end-to-end time series model that aims to learn the such inference relation and make a multiple-step ahead forecast. Our framework trains jointly two neural networks, one to learn the feature-wise correlations and the other for the modeling of temporal behaviors. Our model is capable of simultaneously imputing the missing entries and making a multiple-step ahead prediction. The experiments show good overall performance of our framework over existing methods in both imputation and forecasting tasks.
☆ In-Context Learning User Simulators for Task-Oriented Dialog Systems
This paper presents a novel application of large language models in user simulation for task-oriented dialog systems, specifically focusing on an in-context learning approach. By harnessing the power of these models, the proposed approach generates diverse utterances based on user goals and limited dialog examples. Unlike traditional simulators, this method eliminates the need for labor-intensive rule definition or extensive annotated data, making it more efficient and accessible. Additionally, an error analysis of the interaction between the user simulator and dialog system uncovers common mistakes, providing valuable insights into areas that require improvement. Our implementation is available at https://github.com/telepathylabsai/prompt-based-user-simulator.
☆ Efficient Failure Pattern Identification of Predictive Algorithms UAI2023
Given a (machine learning) classifier and a collection of unlabeled data, how can we efficiently identify misclassification patterns presented in this dataset? To address this problem, we propose a human-machine collaborative framework that consists of a team of human annotators and a sequential recommendation algorithm. The recommendation algorithm is conceptualized as a stochastic sampler that, in each round, queries the annotators a subset of samples for their true labels and obtains the feedback information on whether the samples are misclassified. The sampling mechanism needs to balance between discovering new patterns of misclassification (exploration) and confirming the potential patterns of classification (exploitation). We construct a determinantal point process, whose intensity balances the exploration-exploitation trade-off through the weighted update of the posterior at each round to form the generator of the stochastic sampler. The numerical results empirically demonstrate the competitive performance of our framework on multiple datasets at various signal-to-noise ratios.
comment: 19 pages, Accepted for UAI2023
☆ LiT-4-RSVQA: Lightweight Transformer-based Visual Question Answering in Remote Sensing
Visual question answering (VQA) methods in remote sensing (RS) aim to answer natural language questions with respect to an RS image. Most of the existing methods require a large amount of computational resources, which limits their application in operational scenarios in RS. To address this issue, in this paper we present an effective lightweight transformer-based VQA in RS (LiT-4-RSVQA) architecture for efficient and accurate VQA in RS. Our architecture consists of: i) a lightweight text encoder module; ii) a lightweight image encoder module; iii) a fusion module; and iv) a classification module. The experimental results obtained on a VQA benchmark dataset demonstrate that our proposed LiT-4-RSVQA architecture provides accurate VQA results while significantly reducing the computational requirements on the executing hardware. Our code is publicly available at https://git.tu-berlin.de/rsim/lit4rsvqa.
comment: Accepted at IEEE International Geoscience and Remote Sensing Symposium 2023
☆ Differentiable Tree Operations Promote Compositional Generalization ICML 2023
In the context of structure-to-structure transformation tasks, learning sequences of discrete symbolic operations poses significant challenges due to their non-differentiability. To facilitate the learning of these symbolic sequences, we introduce a differentiable tree interpreter that compiles high-level symbolic tree operations into subsymbolic matrix operations on tensors. We present a novel Differentiable Tree Machine (DTM) architecture that integrates our interpreter with an external memory and an agent that learns to sequentially select tree operations to execute the target transformation in an end-to-end manner. With respect to out-of-distribution compositional generalization on synthetic semantic parsing and language generation tasks, DTM achieves 100% while existing baselines such as Transformer, Tree Transformer, LSTM, and Tree2Tree LSTM achieve less than 30%. DTM remains highly interpretable in addition to its perfect performance.
comment: ICML 2023. Code available at https://github.com/psoulos/dtm
☆ End-to-End Document Classification and Key Information Extraction using Assignment Optimization
We propose end-to-end document classification and key information extraction (KIE) for automating document processing in forms. Through accurate document classification we harness known information from templates to enhance KIE from forms. We use text and layout encoding with a cosine similarity measure to classify visually-similar documents. We then demonstrate a novel application of mixed integer programming by using assignment optimization to extract key information from documents. Our approach is validated on an in-house dataset of noisy scanned forms. The best performing document classification approach achieved 0.97 f1 score. A mean f1 score of 0.94 for the KIE task suggests there is significant potential in applying optimization techniques. Abation results show that the method relies on document preprocessing techniques to mitigate Type II errors and achieve optimal performance.
comment: 10 pages, 5 figures
☆ Going Deeper with Spectral Embeddings
To make sense of millions of raw data and represent them efficiently, practitioners rely on representation learning. Recently, deep connections have been shown between these approaches and the spectral decompositions of some underlying operators. Historically, explicit spectral embeddings were built from graphs constructed on top of the data. In contrast, we propose two new methods to build spectral embeddings: one based on functional analysis principles and kernel methods, which leads to algorithms with theoretical guarantees, and the other based on deep networks trained to optimize principled variational losses, which yield practically efficient algorithms. Furthermore, we provide a new sampling algorithm that leverages learned representations to generate new samples in a single step.
☆ A Uniform Confidence Phenomenon in Deep Learning and its Implications for Calibration
Despite the impressive generalization capabilities of deep neural networks, they have been repeatedly shown to poorly estimate their predictive uncertainty - in other words, they are frequently overconfident when they are wrong. Fixing this issue is known as model calibration, and has consequently received much attention in the form of modified training schemes and post-training calibration procedures. In this work, we present a significant hurdle to the calibration of modern models: deep neural networks have large neighborhoods of almost certain confidence around their training points. We demonstrate in our experiments that this phenomenon consistently arises (in the context of image classification) across many model and dataset pairs. Furthermore, we prove that when this phenomenon holds, for a large class of data distributions with overlaps between classes, it is not possible to obtain a model that is asymptotically better than random (with respect to calibration) even after applying the standard post-training calibration technique of temperature scaling. On the other hand, we also prove that it is possible to circumvent this defect by changing the training process to use a modified loss based on the Mixup data augmentation technique.
comment: 23 pages, 13 Figures
☆ Sharper Bounds for $\ell_p$ Sensitivity Sampling ICML 2023
In large scale machine learning, random sampling is a popular way to approximate datasets by a small representative subset of examples. In particular, sensitivity sampling is an intensely studied technique which provides provable guarantees on the quality of approximation, while reducing the number of examples to the product of the VC dimension $d$ and the total sensitivity $\mathfrak S$ in remarkably general settings. However, guarantees going beyond this general bound of $\mathfrak S d$ are known in perhaps only one setting, for $\ell_2$ subspace embeddings, despite intense study of sensitivity sampling in prior work. In this work, we show the first bounds for sensitivity sampling for $\ell_p$ subspace embeddings for $p\neq 2$ that improve over the general $\mathfrak S d$ bound, achieving a bound of roughly $\mathfrak S^{2/p}$ for $1\leq p<2$ and $\mathfrak S^{2-2/p}$ for $2
comment: To appear in ICML 2023
☆ Dissecting Arbitrary-scale Super-resolution Capability from Pre-trained Diffusion Generative Models
Diffusion-based Generative Models (DGMs) have achieved unparalleled performance in synthesizing high-quality visual content, opening up the opportunity to improve image super-resolution (SR) tasks. Recent solutions for these tasks often train architecture-specific DGMs from scratch, or require iterative fine-tuning and distillation on pre-trained DGMs, both of which take considerable time and hardware investments. More seriously, since the DGMs are established with a discrete pre-defined upsampling scale, they cannot well match the emerging requirements of arbitrary-scale super-resolution (ASSR), where a unified model adapts to arbitrary upsampling scales, instead of preparing a series of distinct models for each case. These limitations beg an intriguing question: can we identify the ASSR capability of existing pre-trained DGMs without the need for distillation or fine-tuning? In this paper, we take a step towards resolving this matter by proposing Diff-SR, a first ASSR attempt based solely on pre-trained DGMs, without additional training efforts. It is motivated by an exciting finding that a simple methodology, which first injects a specific amount of noise into the low-resolution images before invoking a DGM's backward diffusion process, outperforms current leading solutions. The key insight is determining a suitable amount of noise to inject, i.e., small amounts lead to poor low-level fidelity, while over-large amounts degrade the high-level signature. Through a finely-grained theoretical analysis, we propose the Perceptual Recoverable Field (PRF), a metric that achieves the optimal trade-off between these two factors. Extensive experiments verify the effectiveness, flexibility, and adaptability of Diff-SR, demonstrating superior performance to state-of-the-art solutions under diverse ASSR environments.
☆ Renormalized Graph Neural Networks
Graph Neural Networks (GNNs) have become essential for studying complex data, particularly when represented as graphs. Their value is underpinned by their ability to reflect the intricacies of numerous areas, ranging from social to biological networks. GNNs can grapple with non-linear behaviors, emerging patterns, and complex connections; these are also typical characteristics of complex systems. The renormalization group (RG) theory has emerged as the language for studying complex systems. It is recognized as the preferred lens through which to study complex systems, offering a framework that can untangle their intricate dynamics. Despite the clear benefits of integrating RG theory with GNNs, no existing methods have ventured into this promising territory. This paper proposes a new approach that applies RG theory to devise a novel graph rewiring to improve GNNs' performance on graph-related tasks. We support our proposal with extensive experiments on standard benchmarks and baselines. The results demonstrate the effectiveness of our method and its potential to remedy the current limitations of GNNs. Finally, this paper marks the beginning of a new research direction. This path combines the theoretical foundations of RG, the magnifying glass of complex systems, with the structural capabilities of GNNs. By doing so, we aim to enhance the potential of GNNs in modeling and unraveling the complexities inherent in diverse systems.
☆ Spreads in Effective Learning Rates: The Perils of Batch Normalization During Early Training
Excursions in gradient magnitude pose a persistent challenge when training deep networks. In this paper, we study the early training phases of deep normalized ReLU networks, accounting for the induced scale invariance by examining effective learning rates (LRs). Starting with the well-known fact that batch normalization (BN) leads to exponentially exploding gradients at initialization, we develop an ODE-based model to describe early training dynamics. Our model predicts that in the gradient flow, effective LRs will eventually equalize, aligning with empirical findings on warm-up training. Using large LRs is analogous to applying an explicit solver to a stiff non-linear ODE, causing overshooting and vanishing gradients in lower layers after the first step. Achieving overall balance demands careful tuning of LRs, depth, and (optionally) momentum. Our model predicts the formation of spreads in effective LRs, consistent with empirical measurements. Moreover, we observe that large spreads in effective LRs result in training issues concerning accuracy, indicating the importance of controlling these dynamics. To further support a causal relationship, we implement a simple scheduling scheme prescribing uniform effective LRs across layers and confirm accuracy benefits.
☆ Prediction of Post-Operative Renal and Pulmonary Complication Using Transformers
Postoperative complications pose a significant challenge in the healthcare industry, resulting in elevated healthcare expenses and prolonged hospital stays, and in rare instances, patient mortality. To improve patient outcomes and reduce healthcare costs, healthcare providers rely on various perioperative risk scores to guide clinical decisions and prioritize care. In recent years, machine learning techniques have shown promise in predicting postoperative complications and fatality, with deep learning models achieving remarkable success in healthcare applications. However, research on the application of deep learning models to intra-operative anesthesia management data is limited. In this paper, we evaluate the performance of transformer-based models in predicting postoperative acute renal failure, postoperative pulmonary complications, and postoperative in-hospital mortality. We compare our method's performance with state-of-the-art tabular data prediction models, including gradient boosting trees and sequential attention models, on a clinical dataset. Our results demonstrate that transformer-based models can achieve superior performance in predicting postoperative complications and outperform traditional machine learning models. This work highlights the potential of deep learning techniques, specifically transformer-based models, in revolutionizing the healthcare industry's approach to postoperative care.
☆ Stuttering Detection Using Speaker Representations and Self-supervised Contextual Embeddings
The adoption of advanced deep learning architectures in stuttering detection (SD) tasks is challenging due to the limited size of the available datasets. To this end, this work introduces the application of speech embeddings extracted from pre-trained deep learning models trained on large audio datasets for different tasks. In particular, we explore audio representations obtained using emphasized channel attention, propagation, and aggregation time delay neural network (ECAPA-TDNN) and Wav2Vec2.0 models trained on VoxCeleb and LibriSpeech datasets respectively. After extracting the embeddings, we benchmark with several traditional classifiers, such as the K-nearest neighbour (KNN), Gaussian naive Bayes, and neural network, for the SD tasks. In comparison to the standard SD systems trained only on the limited SEP-28k dataset, we obtain a relative improvement of 12.08%, 28.71%, 37.9% in terms of unweighted average recall (UAR) over the baselines. Finally, we have shown that combining two embeddings and concatenating multiple layers of Wav2Vec2.0 can further improve the UAR by up to 2.60% and 6.32% respectively.
comment: Accepted in International Journal of Speech Technology, Springer 2023 substantial overlap with arXiv:2204.01564
☆ Adversarial Robustness in Unsupervised Machine Learning: A Systematic Review
As the adoption of machine learning models increases, ensuring robust models against adversarial attacks is increasingly important. With unsupervised machine learning gaining more attention, ensuring it is robust against attacks is vital. This paper conducts a systematic literature review on the robustness of unsupervised learning, collecting 86 papers. Our results show that most research focuses on privacy attacks, which have effective defenses; however, many attacks lack effective and general defensive measures. Based on the results, we formulate a model on the properties of an attack on unsupervised learning, contributing to future research by providing a model to use.
comment: 38 pages, 11 figures
☆ Balanced Training of Energy-Based Models with Adaptive Flow Sampling
Energy-based models (EBMs) are versatile density estimation models that directly parameterize an unnormalized log density. Although very flexible, EBMs lack a specified normalization constant of the model, making the likelihood of the model computationally intractable. Several approximate samplers and variational inference techniques have been proposed to estimate the likelihood gradients for training. These techniques have shown promising results in generating samples, but little attention has been paid to the statistical accuracy of the estimated density, such as determining the relative importance of different classes in a dataset. In this work, we propose a new maximum likelihood training algorithm for EBMs that uses a different type of generative model, normalizing flows (NF), which have recently been proposed to facilitate sampling. Our method fits an NF to an EBM during training so that an NF-assisted sampling scheme provides an accurate gradient for the EBMs at all times, ultimately leading to a fast sampler for generating new data.
☆ RHFedMTL: Resource-Aware Hierarchical Federated Multi-Task Learning
The rapid development of artificial intelligence (AI) over massive applications including Internet-of-things on cellular network raises the concern of technical challenges such as privacy, heterogeneity and resource efficiency. Federated learning is an effective way to enable AI over massive distributed nodes with security. However, conventional works mostly focus on learning a single global model for a unique task across the network, and are generally less competent to handle multi-task learning (MTL) scenarios with stragglers at the expense of acceptable computation and communication cost. Meanwhile, it is challenging to ensure the privacy while maintain a coupled multi-task learning across multiple base stations (BSs) and terminals. In this paper, inspired by the natural cloud-BS-terminal hierarchy of cellular works, we provide a viable resource-aware hierarchical federated MTL (RHFedMTL) solution to meet the heterogeneity of tasks, by solving different tasks within the BSs and aggregating the multi-task result in the cloud without compromising the privacy. Specifically, a primal-dual method has been leveraged to effectively transform the coupled MTL into some local optimization sub-problems within BSs. Furthermore, compared with existing methods to reduce resource cost by simply changing the aggregation frequency, we dive into the intricate relationship between resource consumption and learning accuracy, and develop a resource-aware learning strategy for local terminals and BSs to meet the resource budget. Extensive simulation results demonstrate the effectiveness and superiority of RHFedMTL in terms of improving the learning accuracy and boosting the convergence rate.
comment: 11 pages, 8 figures
☆ CRS-FL: Conditional Random Sampling for Communication-Efficient and Privacy-Preserving Federated Learning
Federated Learning (FL), a privacy-oriented distributed ML paradigm, is being gaining great interest in Internet of Things because of its capability to protect participants data privacy. Studies have been conducted to address challenges existing in standard FL, including communication efficiency and privacy-preserving. But they cannot achieve the goal of making a tradeoff between communication efficiency and model accuracy while guaranteeing privacy. This paper proposes a Conditional Random Sampling (CRS) method and implements it into the standard FL settings (CRS-FL) to tackle the above-mentioned challenges. CRS explores a stochastic coefficient based on Poisson sampling to achieve a higher probability of obtaining zero-gradient unbiasedly, and then decreases the communication overhead effectively without model accuracy degradation. Moreover, we dig out the relaxation Local Differential Privacy (LDP) guarantee conditions of CRS theoretically. Extensive experiment results indicate that (1) in communication efficiency, CRS-FL performs better than the existing methods in metric accuracy per transmission byte without model accuracy reduction in more than 7% sampling ratio (# sampling size / # model size); (2) in privacy-preserving, CRS-FL achieves no accuracy reduction compared with LDP baselines while holding the efficiency, even exceeding them in model accuracy under more sampling ratio conditions.
☆ Attribute-Efficient PAC Learning of Low-Degree Polynomial Threshold Functions with Nasty Noise ICML 2023
The concept class of low-degree polynomial threshold functions (PTFs) plays a fundamental role in machine learning. In this paper, we study PAC learning of $K$-sparse degree-$d$ PTFs on $\mathbb{R}^n$, where any such concept depends only on $K$ out of $n$ attributes of the input. Our main contribution is a new algorithm that runs in time $({nd}/{\epsilon})^{O(d)}$ and under the Gaussian marginal distribution, PAC learns the class up to error rate $\epsilon$ with $O(\frac{K^{4d}}{\epsilon^{2d}} \cdot \log^{5d} n)$ samples even when an $\eta \leq O(\epsilon^d)$ fraction of them are corrupted by the nasty noise of Bshouty et al. (2002), possibly the strongest corruption model. Prior to this work, attribute-efficient robust algorithms are established only for the special case of sparse homogeneous halfspaces. Our key ingredients are: 1) a structural result that translates the attribute sparsity to a sparsity pattern of the Chow vector under the basis of Hermite polynomials, and 2) a novel attribute-efficient robust Chow vector estimation algorithm which uses exclusively a restricted Frobenius norm to either certify a good approximation or to validate a sparsity-induced degree-$2d$ polynomial as a filter to detect corrupted samples.
comment: ICML 2023
☆ Normalization Enhances Generalization in Visual Reinforcement Learning
Recent advances in visual reinforcement learning (RL) have led to impressive success in handling complex tasks. However, these methods have demonstrated limited generalization capability to visual disturbances, which poses a significant challenge for their real-world application and adaptability. Though normalization techniques have demonstrated huge success in supervised and unsupervised learning, their applications in visual RL are still scarce. In this paper, we explore the potential benefits of integrating normalization into visual RL methods with respect to generalization performance. We find that, perhaps surprisingly, incorporating suitable normalization techniques is sufficient to enhance the generalization capabilities, without any additional special design. We utilize the combination of two normalization techniques, CrossNorm and SelfNorm, for generalizable visual RL. Extensive experiments are conducted on DMControl Generalization Benchmark and CARLA to validate the effectiveness of our method. We show that our method significantly improves generalization capability while only marginally affecting sample efficiency. In particular, when integrated with DrQ-v2, our method enhances the test performance of DrQ-v2 on CARLA across various scenarios, from 14% of the training performance to 97%.
☆ Learning Prescriptive ReLU Networks ICML 23
We study the problem of learning optimal policy from a set of discrete treatment options using observational data. We propose a piecewise linear neural network model that can balance strong prescriptive performance and interpretability, which we refer to as the prescriptive ReLU network, or P-ReLU. We show analytically that this model (i) partitions the input space into disjoint polyhedra, where all instances that belong to the same partition receive the same treatment, and (ii) can be converted into an equivalent prescriptive tree with hyperplane splits for interpretability. We demonstrate the flexibility of the P-ReLU network as constraints can be easily incorporated with minor modifications to the architecture. Through experiments, we validate the superior prescriptive accuracy of P-ReLU against competing benchmarks. Lastly, we present examples of interpretable prescriptive trees extracted from trained P-ReLUs using a real-world dataset, for both the unconstrained and constrained scenarios.
comment: 17 pages, 6 figures, accepted at ICML 23
☆ Universal Test-time Adaptation through Weight Ensembling, Diversity Weighting, and Prior Correction
Since distribution shifts are likely to occur during test-time and can drastically decrease the model's performance, online test-time adaptation (TTA) continues to update the model after deployment, leveraging the current test data. Clearly, a method proposed for online TTA has to perform well for all kinds of environmental conditions. By introducing the variable factors 'domain non-stationarity' and 'temporal correlation', we first unfold all practically relevant settings and define the entity as universal TTA. To tackle the problem of universal TTA, we identify and highlight several challenges a self-training based method has to deal with, including: 1) model bias and the occurrence of trivial solutions when performing entropy minimization on varying sequence lengths with and without multiple domain shifts, 2) loss of generalization which exacerbates the adaptation to future domain shifts and the occurrence of catastrophic forgetting, and 3) performance degradation due to shifts in label prior. To prevent the model from becoming biased, we leverage a dataset and model-agnostic certainty and diversity weighting. In order to maintain generalization and prevent catastrophic forgetting, we propose to continually weight-average the source and adapted model. To compensate for disparities in the label prior during test-time, we propose an adaptive additive prior correction scheme. We evaluate our approach, named ROID, on a wide range of settings, datasets, and models, setting new standards in the field of universal TTA.
☆ TriSig: Assessing the statistical significance of triclusters
Tensor data analysis allows researchers to uncover novel patterns and relationships that cannot be obtained from matrix data alone. The information inferred from the patterns provides valuable insights into disease progression, bioproduction processes, weather fluctuations, and group dynamics. However, spurious and redundant patterns hamper this process. This work aims at proposing a statistical frame to assess the probability of patterns in tensor data to deviate from null expectations, extending well-established principles for assessing the statistical significance of patterns in matrix data. A comprehensive discussion on binomial testing for false positive discoveries is entailed at the light of: variable dependencies, temporal dependencies and misalignments, and \textit{p}-value corrections under the Benjamini-Hochberg procedure. Results gathered from the application of state-of-the-art triclustering algorithms over distinct real-world case studies in biochemical and biotechnological domains confer validity to the proposed statistical frame while revealing vulnerabilities of some triclustering searches. The proposed assessment can be incorporated into existing triclustering algorithms to mitigate false positive/spurious discoveries and further prune the search space, reducing their computational complexity. Availability: The code is freely available at https://github.com/JupitersMight/TriSig under the MIT license.
☆ Byzantine-Robust Clustered Federated Learning
This paper focuses on the problem of adversarial attacks from Byzantine machines in a Federated Learning setting where non-Byzantine machines can be partitioned into disjoint clusters. In this setting, non-Byzantine machines in the same cluster have the same underlying data distribution, and different clusters of non-Byzantine machines have different learning tasks. Byzantine machines can adversarially attack any cluster and disturb the training process on clusters they attack. In the presence of Byzantine machines, the goal of our work is to identify cluster membership of non-Byzantine machines and optimize the models learned by each cluster. We adopt the Iterative Federated Clustering Algorithm (IFCA) framework of Ghosh et al. (2020) to alternatively estimate cluster membership and optimize models. In order to make this framework robust against adversarial attacks from Byzantine machines, we use coordinate-wise trimmed mean and coordinate-wise median aggregation methods used by Yin et al. (2018). Specifically, we propose a new Byzantine-Robust Iterative Federated Clustering Algorithm to improve on the results in Ghosh et al. (2019). We prove a convergence rate for this algorithm for strongly convex loss functions. We compare our convergence rate with the convergence rate of an existing algorithm, and we demonstrate the performance of our algorithm on simulated data.
☆ Unfair Utilities and First Steps Towards Improving Them
Many fairness criteria constrain the policy or choice of predictors. In this work, we propose a different framework for thinking about fairness: Instead of constraining the policy or choice of predictors, we consider which utility a policy is optimizing for. We define value of information fairness and propose to not use utilities that do not satisfy this criterion. We describe how to modify a utility to satisfy this fairness criterion and discuss the consequences this might have on the corresponding optimal policies.
comment: 20 pages
☆ Class Anchor Margin Loss for Content-Based Image Retrieval
The performance of neural networks in content-based image retrieval (CBIR) is highly influenced by the chosen loss (objective) function. The majority of objective functions for neural models can be divided into metric learning and statistical learning. Metric learning approaches require a pair mining strategy that often lacks efficiency, while statistical learning approaches are not generating highly compact features due to their indirect feature optimization. To this end, we propose a novel repeller-attractor loss that falls in the metric learning paradigm, yet directly optimizes for the L2 metric without the need of generating pairs. Our loss is formed of three components. One leading objective ensures that the learned features are attracted to each designated learnable class anchor. The second loss component regulates the anchors and forces them to be separable by a margin, while the third objective ensures that the anchors do not collapse to zero. Furthermore, we develop a more efficient two-stage retrieval system by harnessing the learned class anchors during the first stage of the retrieval process, eliminating the need of comparing the query with every image in the database. We establish a set of four datasets (CIFAR-100, Food-101, SVHN, and Tiny ImageNet) and evaluate the proposed objective in the context of few-shot and full-set training on the CBIR task, by using both convolutional and transformer architectures. Compared to existing objective functions, our empirical evidence shows that the proposed objective is generating superior and more consistent results.
☆ Identifiability and Generalizability in Constrained Inverse Reinforcement Learning ICML 2023
Two main challenges in Reinforcement Learning (RL) are designing appropriate reward functions and ensuring the safety of the learned policy. To address these challenges, we present a theoretical framework for Inverse Reinforcement Learning (IRL) in constrained Markov decision processes. From a convex-analytic perspective, we extend prior results on reward identifiability and generalizability to both the constrained setting and a more general class of regularizations. In particular, we show that identifiability up to potential shaping (Cao et al., 2021) is a consequence of entropy regularization and may generally no longer hold for other regularizations or in the presence of safety constraints. We also show that to ensure generalizability to new transition laws and constraints, the true reward must be identified up to a constant. Additionally, we derive a finite sample guarantee for the suboptimality of the learned rewards, and validate our results in a gridworld environment.
comment: Published at ICML 2023
☆ From Temporal to Contemporaneous Iterative Causal Discovery in the Presence of Latent Confounders ICML
We present a constraint-based algorithm for learning causal structures from observational time-series data, in the presence of latent confounders. We assume a discrete-time, stationary structural vector autoregressive process, with both temporal and contemporaneous causal relations. One may ask if temporal and contemporaneous relations should be treated differently. The presented algorithm gradually refines a causal graph by learning long-term temporal relations before short-term ones, where contemporaneous relations are learned last. This ordering of causal relations to be learnt leads to a reduction in the required number of statistical tests. We validate this reduction empirically and demonstrate that it leads to higher accuracy for synthetic data and more plausible causal graphs for real-world data compared to state-of-the-art algorithms.
comment: Proceedings of the 40-th International Conference on Machine Learning (ICML), 2023
☆ OTW: Optimal Transport Warping for Time Series ICASSP 2023
Dynamic Time Warping (DTW) has become the pragmatic choice for measuring distance between time series. However, it suffers from unavoidable quadratic time complexity when the optimal alignment matrix needs to be computed exactly. This hinders its use in deep learning architectures, where layers involving DTW computations cause severe bottlenecks. To alleviate these issues, we introduce a new metric for time series data based on the Optimal Transport (OT) framework, called Optimal Transport Warping (OTW). OTW enjoys linear time/space complexity, is differentiable and can be parallelized. OTW enjoys a moderate sensitivity to time and shape distortions, making it ideal for time series. We show the efficacy and efficiency of OTW on 1-Nearest Neighbor Classification and Hierarchical Clustering, as well as in the case of using OTW instead of DTW in Deep Learning architectures.
comment: This is an extended version of an ICASSP 2023 accepted paper https://ieeexplore.ieee.org/document/10095915
☆ Effective Structured Prompting by Meta-Learning and Representative Verbalizer ICML 2023
Prompt tuning for pre-trained masked language models (MLM) has shown promising performance in natural language processing tasks with few labeled examples. It tunes a prompt for the downstream task, and a verbalizer is used to bridge the predicted token and label prediction. Due to the limited training data, prompt initialization is crucial for prompt tuning. Recently, MetaPrompting (Hou et al., 2022) uses meta-learning to learn a shared initialization for all task-specific prompts. However, a single initialization is insufficient to obtain good prompts for all tasks and samples when the tasks are complex. Moreover, MetaPrompting requires tuning the whole MLM, causing a heavy burden on computation and memory as the MLM is usually large. To address these issues, we use a prompt pool to extract more task knowledge and construct instance-dependent prompts via attention. We further propose a novel soft verbalizer (RepVerb) which constructs label embedding from feature embeddings directly. Combining meta-learning the prompt pool and RepVerb, we propose MetaPrompter for effective structured prompting. MetaPrompter is parameter-efficient as only the pool is required to be tuned. Experimental results demonstrate that MetaPrompter performs better than the recent state-of-the-arts and RepVerb outperforms existing soft verbalizers.
comment: Accepted at ICML 2023
☆ Progressive Learning for Physics-informed Neural Motion Planning
Motion planning (MP) is one of the core robotics problems requiring fast methods for finding a collision-free robot motion path connecting the given start and goal states. Neural motion planners (NMPs) demonstrate fast computational speed in finding path solutions but require a huge amount of expert trajectories for learning, thus adding a significant training computational load. In contrast, recent advancements have also led to a physics-informed NMP approach that directly solves the Eikonal equation for motion planning and does not require expert demonstrations for learning. However, experiments show that the physics-informed NMP approach performs poorly in complex environments and lacks scalability in multiple scenarios and high-dimensional real robot settings. To overcome these limitations, this paper presents a novel and tractable Eikonal equation formulation and introduces a new progressive learning strategy to train neural networks without expert data in complex, cluttered, multiple high-dimensional robot motion planning scenarios. The results demonstrate that our method outperforms state-of-the-art traditional MP, data-driven NMP, and physics-informed NMP methods by a significant margin in terms of computational planning speed, path quality, and success rates. We also show that our approach scales to multiple complex, cluttered scenarios and the real robot set up in a narrow passage environment. The proposed method's videos and code implementations are available at https://github.com/ruiqini/P-NTFields.
comment: Accepted to Robotics: Science and Systems (RSS) 2023
☆ Adaptation and Optimization of Automatic Speech Recognition (ASR) for the Maritime Domain in the Field of VHF Communication
This paper introduces a multilingual automatic speech recognizer (ASR) for maritime radio communi-cation that automatically converts received VHF radio signals into text. The challenges of maritime radio communication are described at first, and the deep learning architecture of marFM consisting of audio processing techniques and machine learning algorithms is presented. Subsequently, maritime radio data of interest is analyzed and then used to evaluate the transcription performance of our ASR model for various maritime radio data.
☆ On the Effectiveness of Hybrid Mutual Information Estimation
Estimating the mutual information from samples from a joint distribution is a challenging problem in both science and engineering. In this work, we realize a variational bound that generalizes both discriminative and generative approaches. Using this bound, we propose a hybrid method to mitigate their respective shortcomings. Further, we propose Predictive Quantization (PQ): a simple generative method that can be easily combined with discriminative estimators for minimal computational overhead. Our propositions yield a tighter bound on the information thanks to the reduced variance of the estimator. We test our methods on a challenging task of correlated high-dimensional Gaussian distributions and a stochastic process involving a system of free particles subjected to a fixed energy landscape. Empirical results show that hybrid methods consistently improved mutual information estimates when compared to the corresponding discriminative counterpart.
☆ FACT: Federated Adversarial Cross Training
Federated Learning (FL) facilitates distributed model development to aggregate multiple confidential data sources. The information transfer among clients can be compromised by distributional differences, i.e., by non-i.i.d. data. A particularly challenging scenario is the federated model adaptation to a target client without access to annotated data. We propose Federated Adversarial Cross Training (FACT), which uses the implicit domain differences between source clients to identify domain shifts in the target domain. In each round of FL, FACT cross initializes a pair of source clients to generate domain specialized representations which are then used as a direct adversary to learn a domain invariant data representation. We empirically show that FACT outperforms state-of-the-art federated, non-federated and source-free domain adaptation models on three popular multi-source-single-target benchmarks, and state-of-the-art Unsupervised Domain Adaptation (UDA) models on single-source-single-target experiments. We further study FACT's behavior with respect to communication restrictions and the number of participating clients.
☆ Safe Offline Reinforcement Learning with Real-Time Budget Constraints
Aiming at promoting the safe real-world deployment of Reinforcement Learning (RL), research on safe RL has made significant progress in recent years. However, most existing works in the literature still focus on the online setting where risky violations of the safety budget are likely to be incurred during training. Besides, in many real-world applications, the learned policy is required to respond to dynamically determined safety budgets (i.e., constraint threshold) in real time. In this paper, we target at the above real-time budget constraint problem under the offline setting, and propose Trajectory-based REal-time Budget Inference (TREBI) as a novel solution that approaches this problem from the perspective of trajectory distribution. Theoretically, we prove an error bound of the estimation on the episodic reward and cost under the offline setting and thus provide a performance guarantee for TREBI. Empirical results on a wide range of simulation tasks and a real-world large-scale advertising application demonstrate the capability of TREBI in solving real-time budget constraint problems under offline settings.
comment: We propose a method to handle the constraint problem with dynamically determined safety budgets under the offline setting
☆ Approximate Stein Classes for Truncated Density Estimation ICML 2023
Estimating truncated density models is difficult, as these models have intractable normalising constants and hard to satisfy boundary conditions. Score matching can be adapted to solve the truncated density estimation problem, but requires a continuous weighting function which takes zero at the boundary and is positive elsewhere. Evaluation of such a weighting function (and its gradient) often requires a closed-form expression of the truncation boundary and finding a solution to a complicated optimisation problem. In this paper, we propose approximate Stein classes, which in turn leads to a relaxed Stein identity for truncated density estimation. We develop a novel discrepancy measure, truncated kernelised Stein discrepancy (TKSD), which does not require fixing a weighting function in advance, and can be evaluated using only samples on the boundary. We estimate a truncated density model by minimising the Lagrangian dual of TKSD. Finally, experiments show the accuracy of our method to be an improvement over previous works even without the explicit functional form of the boundary.
comment: Accepted to ICML 2023
☆ Rotating Features for Object Discovery
The binding problem in human cognition, concerning how the brain represents and connects objects within a fixed network of neural connections, remains a subject of intense debate. Most machine learning efforts addressing this issue in an unsupervised setting have focused on slot-based methods, which may be limiting due to their discrete nature and difficulty to express uncertainty. Recently, the Complex AutoEncoder was proposed as an alternative that learns continuous and distributed object-centric representations. However, it is only applicable to simple toy data. In this paper, we present Rotating Features, a generalization of complex-valued features to higher dimensions, and a new evaluation procedure for extracting objects from distributed representations. Additionally, we show the applicability of our approach to pre-trained features. Together, these advancements enable us to scale distributed object-centric representations from simple toy to real-world data. We believe this work advances a new paradigm for addressing the binding problem in machine learning and has the potential to inspire further innovation in the field.
☆ Evaluating the "Learning on Graphs" Conference Experience
With machine learning conferences growing ever larger, and reviewing processes becoming increasingly elaborate, more data-driven insights into their workings are required. In this report, we present the results of a survey accompanying the first "Learning on Graphs" (LoG) Conference. The survey was directed to evaluate the submission and review process from different perspectives, including authors, reviewers, and area chairs alike.
☆ Causal Imitability Under Context-Specific Independence Relations
Drawbacks of ignoring the causal mechanisms when performing imitation learning have recently been acknowledged. Several approaches both to assess the feasibility of imitation and to circumvent causal confounding and causal misspecifications have been proposed in the literature. However, the potential benefits of the incorporation of additional information about the underlying causal structure are left unexplored. An example of such overlooked information is context-specific independence (CSI), i.e., independence that holds only in certain contexts. We consider the problem of causal imitation learning when CSI relations are known. We prove that the decision problem pertaining to the feasibility of imitation in this setting is NP-hard. Further, we provide a necessary graphical criterion for imitation learning under CSI and show that under a structural assumption, this criterion is also sufficient. Finally, we propose a sound algorithmic approach for causal imitation learning which takes both CSI relations and data into account.
comment: 19 pages, 4 figures, under review
☆ Anomaly Detection with Variance Stabilized Density Estimation
Density estimation based anomaly detection schemes typically model anomalies as examples that reside in low-density regions. We propose a modified density estimation problem and demonstrate its effectiveness for anomaly detection. Specifically, we assume the density function of normal samples is uniform in some compact domain. This assumption implies the density function is more stable (with lower variance) around normal samples than anomalies. We first corroborate this assumption empirically using a wide range of real-world data. Then, we design a variance stabilized density estimation problem for maximizing the likelihood of the observed samples while minimizing the variance of the density around normal samples. We introduce an ensemble of autoregressive models to learn the variance stabilized distribution. Finally, we perform an extensive benchmark with 52 datasets demonstrating that our method leads to state-of-the-art results while alleviating the need for data-specific hyperparameter tuning.
comment: 12 pages, 6 figures
☆ Does Black-box Attribute Inference Attacks on Graph Neural Networks Constitute Privacy Risk?
Graph neural networks (GNNs) have shown promising results on real-life datasets and applications, including healthcare, finance, and education. However, recent studies have shown that GNNs are highly vulnerable to attacks such as membership inference attack and link reconstruction attack. Surprisingly, attribute inference attacks has received little attention. In this paper, we initiate the first investigation into attribute inference attack where an attacker aims to infer the sensitive user attributes based on her public or non-sensitive attributes. We ask the question whether black-box attribute inference attack constitutes a significant privacy risk for graph-structured data and their corresponding GNN model. We take a systematic approach to launch the attacks by varying the adversarial knowledge and assumptions. Our findings reveal that when an attacker has black-box access to the target model, GNNs generally do not reveal significantly more information compared to missing value estimation techniques. Code is available.
☆ TorchRL: A data-driven decision-making library for PyTorch
Striking a balance between integration and modularity is crucial for a machine learning library to be versatile and user-friendly, especially in handling decision and control tasks that involve large development teams and complex, real-world data, and environments. To address this issue, we propose TorchRL, a generalistic control library for PyTorch that provides well-integrated, yet standalone components. With a versatile and robust primitive design, TorchRL facilitates streamlined algorithm development across the many branches of Reinforcement Learning (RL) and control. We introduce a new PyTorch primitive, TensorDict, as a flexible data carrier that empowers the integration of the library's components while preserving their modularity. Hence replay buffers, datasets, distributed data collectors, environments, transforms and objectives can be effortlessly used in isolation or combined. We provide a detailed description of the building blocks, supporting code examples and an extensive overview of the library across domains and tasks. Finally, we show comparative benchmarks to demonstrate its computational efficiency. TorchRL fosters long-term support and is publicly available on GitHub for greater reproducibility and collaboration within the research community. The code is opensourced on https://github.com/pytorch/rl.
☆ Predicting Temporal Aspects of Movement for Predictive Replication in Fog Environments
To fully exploit the benefits of the fog environment, efficient management of data locality is crucial. Blind or reactive data replication falls short in harnessing the potential of fog computing, necessitating more advanced techniques for predicting where and when clients will connect. While spatial prediction has received considerable attention, temporal prediction remains understudied. Our paper addresses this gap by examining the advantages of incorporating temporal prediction into existing spatial prediction models. We also provide a comprehensive analysis of spatio-temporal prediction models, such as Deep Neural Networks and Markov models, in the context of predictive replication. We propose a novel model using Holt-Winter's Exponential Smoothing for temporal prediction, leveraging sequential and periodical user movement patterns. In a fog network simulation with real user trajectories our model achieves a 15% reduction in excess data with a marginal 1% decrease in data availability.
☆ Machine Learning and Kalman Filtering for Nanomechanical Mass Spectrometry
Nanomechanical resonant sensors are used in mass spectrometry via detection of resonance frequency jumps. There is a fundamental trade-off between detection speed and accuracy. Temporal and size resolution are limited by the resonator characteristics and noise. A Kalman filtering technique, augmented with maximum-likelihood estimation, was recently proposed as a Pareto optimal solution. We present enhancements and robust realizations for this technique, including a confidence boosted thresholding approach as well as machine learning for event detection. We describe learning techniques that are based on neural networks and boosted decision trees for temporal location and event size estimation. In the pure learning based approach that discards the Kalman filter, the raw data from the sensor are used in training a model for both location and size prediction. In the alternative approach that augments a Kalman filter, the event likelihood history is used in a binary classifier for event occurrence. Locations and sizes are predicted using maximum-likelihood, followed by a Kalman filter that continually improves the size estimate. We present detailed comparisons of the learning based schemes and the confidence boosted thresholding approach, and demonstrate robust performance for a practical realization.
comment: 8 pages, 9 figures
☆ Hinge-Wasserstein: Mitigating Overconfidence in Regression by Classification
Modern deep neural networks are prone to being overconfident despite their drastically improved performance. In ambiguous or even unpredictable real-world scenarios, this overconfidence can pose a major risk to the safety of applications. For regression tasks, the regression-by-classification approach has the potential to alleviate these ambiguities by instead predicting a discrete probability density over the desired output. However, a density estimator still tends to be overconfident when trained with the common NLL loss. To mitigate the overconfidence problem, we propose a loss function, hinge-Wasserstein, based on the Wasserstein Distance. This loss significantly improves the quality of both aleatoric and epistemic uncertainty, compared to previous work. We demonstrate the capabilities of the new loss on a synthetic dataset, where both types of uncertainty are controlled separately. Moreover, as a demonstration for real-world scenarios, we evaluate our approach on the benchmark dataset Horizon Lines in the Wild. On this benchmark, using the hinge-Wasserstein loss reduces the Area Under Sparsification Error (AUSE) for horizon parameters slope and offset, by 30.47% and 65.00%, respectively.
☆ We never go out of Style: Motion Disentanglement by Subspace Decomposition of Latent Space CVPR
Real-world objects perform complex motions that involve multiple independent motion components. For example, while talking, a person continuously changes their expressions, head, and body pose. In this work, we propose a novel method to decompose motion in videos by using a pretrained image GAN model. We discover disentangled motion subspaces in the latent space of widely used style-based GAN models that are semantically meaningful and control a single explainable motion component. The proposed method uses only a few $(\approx10)$ ground truth video sequences to obtain such subspaces. We extensively evaluate the disentanglement properties of motion subspaces on face and car datasets, quantitatively and qualitatively. Further, we present results for multiple downstream tasks such as motion editing, and selective motion transfer, e.g. transferring only facial expressions without training for it.
comment: AI for content creation, CVPRW-2023
♻ ☆ LFTK: Handcrafted Features in Computational Linguistics ACL 2023
Past research has identified a rich set of handcrafted linguistic features that can potentially assist various tasks. However, their extensive number makes it difficult to effectively select and utilize existing handcrafted features. Coupled with the problem of inconsistent implementation across research works, there has been no categorization scheme or generally-accepted feature names. This creates unwanted confusion. Also, most existing handcrafted feature extraction libraries are not open-source or not actively maintained. As a result, a researcher often has to build such an extraction system from the ground up. We collect and categorize more than 220 popular handcrafted features grounded on past literature. Then, we conduct a correlation analysis study on several task-specific datasets and report the potential use cases of each feature. Lastly, we devise a multilingual handcrafted linguistic feature extraction system in a systematically expandable manner. We open-source our system for public access to a rich set of pre-implemented handcrafted features. Our system is coined LFTK and is the largest of its kind. Find it at github.com/brucewlee/lftk.
comment: BEA @ ACL 2023
♻ ☆ Towards Revealing the Mystery behind Chain of Thought: a Theoretical Perspective
Recent studies have discovered that Chain-of-Thought prompting (CoT) can dramatically improve the performance of Large Language Models (LLMs), particularly when dealing with complex tasks involving mathematics or reasoning. Despite the enormous empirical success, the underlying mechanisms behind CoT and how it unlocks the potential of LLMs remain elusive. In this paper, we take a first step towards theoretically answering these questions. Specifically, we examine the expressivity of LLMs with CoT in solving fundamental mathematical and decision-making problems. We start by giving an impossibility result showing that bounded-depth Transformers are unable to directly produce correct answers for basic arithmetic/equation tasks unless the model size grows super-polynomially with respect to the input length. In contrast, we then prove by construction that autoregressive Transformers of constant size suffice to solve both tasks by generating CoT derivations using a commonly-used math language format. Moreover, we show LLMs with CoT are capable of solving a general class of decision-making problems known as Dynamic Programming, thus justifying its power in tackling complex real-world tasks. Finally, extensive experiments on four tasks show that, while Transformers always fail to predict the answers directly, they can consistently learn to generate correct solutions step-by-step given sufficient CoT demonstrations.
comment: 34 pages
♻ ☆ Offline Meta Reinforcement Learning with In-Distribution Online Adaptation
Recent offline meta-reinforcement learning (meta-RL) methods typically utilize task-dependent behavior policies (e.g., training RL agents on each individual task) to collect a multi-task dataset. However, these methods always require extra information for fast adaptation, such as offline context for testing tasks. To address this problem, we first formally characterize a unique challenge in offline meta-RL: transition-reward distribution shift between offline datasets and online adaptation. Our theory finds that out-of-distribution adaptation episodes may lead to unreliable policy evaluation and that online adaptation with in-distribution episodes can ensure adaptation performance guarantee. Based on these theoretical insights, we propose a novel adaptation framework, called In-Distribution online Adaptation with uncertainty Quantification (IDAQ), which generates in-distribution context using a given uncertainty quantification and performs effective task belief inference to address new tasks. We find a return-based uncertainty quantification for IDAQ that performs effectively. Experiments show that IDAQ achieves state-of-the-art performance on the Meta-World ML1 benchmark compared to baselines with/without offline adaptation.
♻ ☆ Federated Conformal Predictors for Distributed Uncertainty Quantification ICML 2023
Conformal prediction is emerging as a popular paradigm for providing rigorous uncertainty quantification in machine learning since it can be easily applied as a post-processing step to already trained models. In this paper, we extend conformal prediction to the federated learning setting. The main challenge we face is data heterogeneity across the clients - this violates the fundamental tenet of exchangeability required for conformal prediction. We propose a weaker notion of partial exchangeability, better suited to the FL setting, and use it to develop the Federated Conformal Prediction (FCP) framework. We show FCP enjoys rigorous theoretical guarantees and excellent empirical performance on several computer vision and medical imaging datasets. Our results demonstrate a practical approach to incorporating meaningful uncertainty quantification in distributed and heterogeneous environments. We provide code used in our experiments https://github.com/clu5/federated-conformal.
comment: 23 pages, 18 figures, accepted to International Conference on Machine Learning (ICML 2023)
♻ ☆ Contrastive Shapelet Learning for Unsupervised Multivariate Time Series Representation Learning
Recent studies have shown great promise in unsupervised representation learning (URL) for multivariate time series, because URL has the capability in learning generalizable representation for many downstream tasks without using inaccessible labels. However, existing approaches usually adopt the models originally designed for other domains (e.g., computer vision) to encode the time series data and rely on strong assumptions to design learning objectives, which limits their ability to perform well. To deal with these problems, we propose a novel URL framework for multivariate time series by learning time-series-specific shapelet-based representation through a popular contrasting learning paradigm. To the best of our knowledge, this is the first work that explores the shapelet-based embedding in the unsupervised general-purpose representation learning. A unified shapelet-based encoder and a novel learning objective with multi-grained contrasting and multi-scale alignment are particularly designed to achieve our goal, and a data augmentation library is employed to improve the generalization. We conduct extensive experiments using tens of real-world datasets to assess the representation quality on many downstream tasks, including classification, clustering, and anomaly detection. The results demonstrate the superiority of our method against not only URL competitors, but also techniques specially designed for downstream tasks. Our code has been made publicly available at https://github.com/real2fish/CSL.
♻ ☆ NN2Poly: A polynomial representation for deep feed-forward artificial neural networks
Interpretability of neural networks and their underlying theoretical behavior remain an open field of study even after the great success of their practical applications, particularly with the emergence of deep learning. In this work, NN2Poly is proposed: a theoretical approach to obtain an explicit polynomial model that provides an accurate representation of an already trained fully-connected feed-forward artificial neural network (a multilayer perceptron or MLP). This approach extends a previous idea proposed in the literature, which was limited to single hidden layer networks, to work with arbitrarily deep MLPs in both regression and classification tasks. The objective of this paper is to achieve this by using a Taylor expansion on the activation function, at each layer, and then using several combinatorial properties to calculate the coefficients of the desired polynomials. Discussion is presented on the main computational challenges of this method, and the way to overcome them by imposing certain constraints during the training phase. Finally, simulation experiments as well as an application to a real data set are presented to demonstrate the effectiveness of the proposed method.
♻ ☆ Heterogeneous Value Evaluation for Large Language Models
The emergent capabilities of Large Language Models (LLMs) have made it crucial to align their values with those of humans. Current methodologies typically attempt alignment with a homogeneous human value and requires human verification, yet lack consensus on the desired aspect and depth of alignment and resulting human biases. In this paper, we propose A2EHV, an Automated Alignment Evaluation with a Heterogeneous Value system that (1) is automated to minimize individual human biases, and (2) allows assessments against various target values to foster heterogeneous agents. Our approach pivots on the concept of value rationality, which represents the ability for agents to execute behaviors that satisfy a target value the most. The quantification of value rationality is facilitated by the Social Value Orientation framework from social psychology, which partitions the value space into four categories to assess social preferences from agents' behaviors. We evaluate the value rationality of eight mainstream LLMs and observe that large models are more inclined to align neutral values compared to those with strong personal values. By examining the behavior of these LLMs, we contribute to a deeper understanding of value alignment within a heterogeneous value system.
comment: Our full prompts are released in the repo: https://github.com/zowiezhang/A2EHV
♻ ☆ A Comprehensive Overview and Comparative Analysis on Deep Learning Models: CNN, RNN, LSTM, GRU
Deep learning (DL) has emerged as a powerful subset of machine learning (ML) and artificial intelligence (AI), outperforming traditional ML methods, especially in handling unstructured and large datasets. Its impact spans across various domains, including speech recognition, healthcare, autonomous vehicles, cybersecurity, predictive analytics, and more. However, the complexity and dynamic nature of real-world problems present challenges in designing effective deep learning models. Consequently, several deep learning models have been developed to address different problems and applications. In this article, we conduct a comprehensive survey of various deep learning models, including Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), Generative Models, Deep Reinforcement Learning (DRL), and Deep Transfer Learning. We examine the structure, applications, benefits, and limitations of each model. Furthermore, we perform an analysis using three publicly available datasets: IMDB, ARAS, and Fruit-360. We compare the performance of six renowned deep learning models: CNN, Simple RNN, Long Short-Term Memory (LSTM), Bidirectional LSTM, Gated Recurrent Unit (GRU), and Bidirectional GRU.
comment: 16 pages, 29 figures
♻ ☆ Sample Complexity of Forecast Aggregation
We consider a Bayesian forecast aggregation model where $n$ experts, after observing private signals about an unknown binary event, report their posterior beliefs about the event to a principal, who then aggregates the reports into a single prediction for the event. The signals of the experts and the outcome of the event follow a joint distribution that is unknown to the principal, but the principal has access to i.i.d. "samples" from the distribution, where each sample is a tuple of the experts' reports (not signals) and the realization of the event. Using these samples, the principal aims to find an $\varepsilon$-approximately optimal aggregator, where optimality is measured in terms of the expected squared distance between the aggregated prediction and the realization of the event. We show that the sample complexity of this problem is at least $\tilde \Omega(m^{n-2} / \varepsilon)$ for arbitrary discrete distributions, where $m$ is the size of each expert's signal space. This sample complexity grows exponentially in the number of experts $n$. But, if the experts' signals are independent conditioned on the realization of the event, then the sample complexity is significantly reduced, to $\tilde O(1 / \varepsilon^2)$, which does not depend on $n$. Our results can be generalized to non-binary events. The proof of our results uses a reduction from the distribution learning problem and reveals the fact that forecast aggregation is almost as difficult as distribution learning.
comment: Update related works. Add new results and discussions
♻ ☆ Continual Vision-Language Representation Learning with Off-Diagonal Information
Large-scale multi-modal contrastive learning frameworks like CLIP typically require a large amount of image-text samples for training. However, these samples are always collected continuously in real scenarios. This paper discusses the feasibility of continual CLIP training using streaming data. Unlike continual learning based on self-supervised learning methods for pure images, which is empirically robust against catastrophic forgetting, CLIP's performance degeneration in the continual setting is significant and non-neglectable. By analyzing the changes in the model's representation space during continual CLIP training from a spatial geometry perspective, we explore and summarize these spatial variations as Spatial Disorder (SD), which can be divided into Intra-modal Rotation and Inter-modal Deviation. Moreover, we empirically and theoretically demonstrate how SD leads to a performance decline for CLIP on cross-modal retrieval tasks. To alleviate SD, we propose a new continual vision-language representation learning framework Mod-X: Maintain off-diagonal information-matriX. By selectively aligning the off-diagonal information distribution of contrastive matrices, the Mod-X improves the capability of the multi-modal model by maintaining the multi-modal representation space alignment on the old data domain during continuously fitting the new training data domain. Experiments on commonly used datasets with different scales and scopes have demonstrated the effectiveness of our method.
♻ ☆ Power Grid Behavioral Patterns and Risks of Generalization in Applied Machine Learning
Recent years have seen a rich literature of data-driven approaches designed for power grid applications. However, insufficient consideration of domain knowledge can impose a high risk to the practicality of the methods. Specifically, ignoring the grid-specific spatiotemporal patterns (in load, generation, and topology, etc.) can lead to outputting infeasible, unrealizable, or completely meaningless predictions on new inputs. To address this concern, this paper investigates real-world operational data to provide insights into power grid behavioral patterns, including the time-varying topology, load, and generation, as well as the spatial differences (in peak hours, diverse styles) between individual loads and generations. Then based on these observations, we evaluate the generalization risks in some existing ML works causedby ignoring these grid-specific patterns in model design and training.
♻ ☆ Arithmetic Sampling: Parallel Diverse Decoding for Large Language Models ICML 2023
Decoding methods for large language models often trade-off between diversity of outputs and parallelism of computation. Methods such as beam search and Gumbel top-k sampling can guarantee a different output for each element of the beam, but are not easy to parallelize. Alternatively, methods such as temperature sampling and its modifications (top-k sampling, nucleus sampling, typical decoding, and others), are embarrassingly parallel, but have no guarantees about duplicate samples. We present a framework for sampling according to an arithmetic code book implicitly defined by a large language model, compatible with common sampling variations, with provable beam diversity under certain conditions, as well as being embarrassingly parallel and providing unbiased and consistent expectations from the original model. We demonstrate the effectiveness of our approach on WMT machine translation, more than halving the standard deviation when estimating expected BLEU score reward, and closing the BLEU score gap between independent sampling and beam search by up to 63%.
comment: 17 pages, to appear at ICML 2023
♻ ☆ Grounding Language Models to Images for Multimodal Inputs and Outputs ICML 2023
We propose an efficient method to ground pretrained text-only language models to the visual domain, enabling them to process arbitrarily interleaved image-and-text data, and generate text interleaved with retrieved images. Our method leverages the abilities of language models learnt from large scale text-only pretraining, such as in-context learning and free-form text generation. We keep the language model frozen, and finetune input and output linear layers to enable cross-modality interactions. This allows our model to process arbitrarily interleaved image-and-text inputs, and generate free-form text interleaved with retrieved images. We achieve strong zero-shot performance on grounded tasks such as contextual image retrieval and multimodal dialogue, and showcase compelling interactive abilities. Our approach works with any off-the-shelf language model and paves the way towards an effective, general solution for leveraging pretrained language models in visually grounded settings.
comment: Published in ICML 2023. Project page: https://jykoh.com/fromage
♻ ☆ Model Transferability With Responsive Decision Subjects
Given an algorithmic predictor that is accurate on some source population consisting of strategic human decision subjects, will it remain accurate if the population respond to it? In our setting, an agent or a user corresponds to a sample $(X,Y)$ drawn from a distribution $\cal{D}$ and will face a model $h$ and its classification result $h(X)$. Agents can modify $X$ to adapt to $h$, which will incur a distribution shift on $(X,Y)$. Our formulation is motivated by applications where the deployed machine learning models are subjected to human agents, and will ultimately face responsive and interactive data distributions. We formalize the discussions of the transferability of a model by studying how the performance of the model trained on the available source distribution (data) would translate to the performance on its induced domain. We provide both upper bounds for the performance gap due to the induced domain shift, as well as lower bounds for the trade-offs that a classifier has to suffer on either the source training distribution or the induced target distribution. We provide further instantiated analysis for two popular domain adaptation settings, including covariate shift and target shift.
♻ ☆ Generalization for slowly mixing processes
A bound uniform over various loss-classes is given for data generated by stationary and phi-mixing processes, where the mixing time (the time needed to obtain approximate independence) enters the sample complexity only in an additive way. For slowly mixing processes this can be a considerable advantage over results with multiplicative dependence on the mixing time. The admissible loss-classes include functions with prescribed Lipschitz norms or smoothness parameters. The bound can also be applied to be uniform over unconstrained loss-classes, where it depends on local Lipschitz properties of the function on the sample path.
comment: Improved version
♻ ☆ Disentangled Causal Graph Learning forOnline Unsupervised Root Cause Analysis
The task of root cause analysis (RCA) is to identify the root causes of system faults/failures by analyzing system monitoring data. Efficient RCA can greatly accelerate system failure recovery and mitigate system damages or financial losses. However, previous research has mostly focused on developing offline RCA algorithms, which often require manually initiating the RCA process, a significant amount of time and data to train a robust model, and then being retrained from scratch for a new system fault. In this paper, we propose CORAL, a novel online RCA framework that can automatically trigger the RCA process and incrementally update the RCA model. CORAL consists of Trigger Point Detection, Incremental Disentangled Causal Graph Learning, and Network Propagation-based Root Cause Localization. The Trigger Point Detection component aims to detect system state transitions automatically and in near-real-time. To achieve this, we develop an online trigger point detection approach based on multivariate singular spectrum analysis and cumulative sum statistics. To efficiently update the RCA model, we propose an incremental disentangled causal graph learning approach to decouple the state-invariant and state-dependent information. After that, CORAL applies a random walk with restarts to the updated causal graph to accurately identify root causes. The online RCA process terminates when the causal graph and the generated root cause list converge. Extensive experiments on three real-world datasets with case studies demonstrate the effectiveness and superiority of the proposed framework.
♻ ☆ Transferable Energy Storage Bidder
Energy storage resources must consider both price uncertainties and their physical operating characteristics when participating in wholesale electricity markets. This is a challenging problem as electricity prices are highly volatile, and energy storage has efficiency losses, power, and energy constraints. This paper presents a novel, versatile, and transferable approach combining model-based optimization with a convolutional long short-term memory network for energy storage to respond to or bid into wholesale electricity markets. We test our proposed approach using historical prices from New York State, showing it achieves state-of-the-art results, achieving between 70% to near 90% profit ratio compared to perfect foresight cases, in both price response and wholesale market bidding setting with various energy storage durations. We also test a transfer learning approach by pre-training the bidding model using New York data and applying it to arbitrage in Queensland, Australia. The result shows transfer learning achieves exceptional arbitrage profitability with as little as three days of local training data, demonstrating its significant advantage over training from scratch in scenarios with very limited data availability.
♻ ☆ A Closer Look at Few-shot Classification Again ICML 2023
Few-shot classification consists of a training phase where a model is learned on a relatively large dataset and an adaptation phase where the learned model is adapted to previously-unseen tasks with limited labeled samples. In this paper, we empirically prove that the training algorithm and the adaptation algorithm can be completely disentangled, which allows algorithm analysis and design to be done individually for each phase. Our meta-analysis for each phase reveals several interesting insights that may help better understand key aspects of few-shot classification and connections with other fields such as visual representation learning and transfer learning. We hope the insights and research challenges revealed in this paper can inspire future work in related directions. Code and pre-trained models (in PyTorch) are available at https://github.com/Frankluox/CloserLookAgainFewShot.
comment: Accepted at ICML 2023
♻ ☆ Learning Runtime Decisions for Adaptive Real-Time Perception
Real-time perception requires planned resource utilization. Computational planning in real-time perception is governed by two considerations -- accuracy and latency. There exist run-time decisions (e.g. choice of input resolution) that induce tradeoffs affecting performance on a given hardware, arising from intrinsic (content, e.g. scene clutter) and extrinsic (system, e.g. resource contention) characteristics. Earlier runtime execution frameworks employed rule-based decision algorithms and operated with a fixed algorithm latency budget to balance these concerns, which is sub-optimal and inflexible. We propose Chanakya, a learned approximate execution framework that naturally derives from the streaming perception paradigm, to automatically learn decisions induced by these tradeoffs instead. Chanakya is trained via novel rewards balancing accuracy and latency implicitly, without approximating either objectives. Chanakya simultaneously considers intrinsic and extrinsic context, and predicts decisions in a flexible manner. Chanakya, designed with low overhead in mind, outperforms state-of-the-art static and dynamic execution policies on public datasets on both server GPUs and edge devices.
comment: 18 pages
♻ ☆ Towards Certification of Machine Learning-Based Distributed Systems
Machine Learning (ML) is increasingly used to drive the operation of complex distributed systems deployed on the cloud-edge continuum enabled by 5G. Correspondingly, distributed systems' behavior is becoming more non-deterministic in nature. This evolution of distributed systems requires the definition of new assurance approaches for the verification of non-functional properties. Certification, the most popular assurance technique for system and software verification, is not immediately applicable to systems whose behavior is determined by Machine Learning-based inference. However, there is an increasing push from policy makers, regulators, and industrial stakeholders towards the definition of techniques for the certification of non-functional properties (e.g., fairness, robustness, privacy) of ML. This article analyzes the challenges and deficiencies of current certification schemes, discusses open research issues and proposes a first certification scheme for ML-based distributed systems.
comment: 5 pages, 1 figure, 1 table
♻ ☆ Graph-based Multi-ODE Neural Networks for Spatio-Temporal Traffic Forecasting
There is a recent surge in the development of spatio-temporal forecasting models in the transportation domain. Long-range traffic forecasting, however, remains a challenging task due to the intricate and extensive spatio-temporal correlations observed in traffic networks. Current works primarily rely on road networks with graph structures and learn representations using graph neural networks (GNNs), but this approach suffers from over-smoothing problem in deep architectures. To tackle this problem, recent methods introduced the combination of GNNs with residual connections or neural ordinary differential equations (ODE). However, current graph ODE models face two key limitations in feature extraction: (1) they lean towards global temporal patterns, overlooking local patterns that are important for unexpected events; and (2) they lack dynamic semantic edges in their architectural design. In this paper, we propose a novel architecture called Graph-based Multi-ODE Neural Networks (GRAM-ODE) which is designed with multiple connective ODE-GNN modules to learn better representations by capturing different views of complex local and global dynamic spatio-temporal dependencies. We also add some techniques like shared weights and divergence constraints into the intermediate layers of distinct ODE-GNN modules to further improve their communication towards the forecasting task. Our extensive set of experiments conducted on six real-world datasets demonstrate the superior performance of GRAM-ODE compared with state-of-the-art baselines as well as the contribution of different components to the overall performance. The code is available at https://github.com/zbliu98/GRAM-ODE
comment: Published in Transactions on Machine Learning Research, 2023
♻ ☆ Efficient Bi-Level Optimization for Recommendation Denoising
The acquisition of explicit user feedback (e.g., ratings) in real-world recommender systems is often hindered by the need for active user involvement. To mitigate this issue, implicit feedback (e.g., clicks) generated during user browsing is exploited as a viable substitute. However, implicit feedback possesses a high degree of noise, which significantly undermines recommendation quality. While many methods have been proposed to address this issue by assigning varying weights to implicit feedback, two shortcomings persist: (1) the weight calculation in these methods is iteration-independent, without considering the influence of weights in previous iterations, and (2) the weight calculation often relies on prior knowledge, which may not always be readily available or universally applicable. To overcome these two limitations, we model recommendation denoising as a bi-level optimization problem. The inner optimization aims to derive an effective model for the recommendation, as well as guiding the weight determination, thereby eliminating the need for prior knowledge. The outer optimization leverages gradients of the inner optimization and adjusts the weights in a manner considering the impact of previous weights. To efficiently solve this bi-level optimization problem, we employ a weight generator to avoid the storage of weights and a one-step gradient-matching-based loss to significantly reduce computational time. The experimental results on three benchmark datasets demonstrate that our proposed approach outperforms both state-of-the-art general and denoising recommendation models. The code is available at https://github.com/CoderWZW/BOD.
comment: 11pages, 5 figures, 6 tables
♻ ☆ Contrastive Multiple Correspondence Analysis (cMCA): Using Contrastive Learning to Identify Latent Subgroups in Political Parties
Scaling methods have long been utilized to simplify and cluster high-dimensional data. However, the general latent spaces across all predefined groups derived from these methods sometimes do not fall into researchers' interest regarding specific patterns within groups. To tackle this issue, we adopt an emerging analysis approach called contrastive learning. We contribute to this growing field by extending its ideas to multiple correspondence analysis (MCA) in order to enable an analysis of data often encountered by social scientists -- containing binary, ordinal, and nominal variables. We demonstrate the utility of contrastive MCA (cMCA) by analyzing two different surveys of voters in the U.S. and U.K. Our results suggest that, first, cMCA can identify substantively important dimensions and divisions among subgroups that are overlooked by traditional methods; second, for other cases, cMCA can derive latent traits that emphasize subgroups seen moderately in those derived by traditional methods.
comment: Both authors contributed equally to this work and are listed alphabetically. This manuscript is accepted by PLOS ONE
♻ ☆ Policy Optimization for Continuous Reinforcement Learning
We study reinforcement learning (RL) in the setting of continuous time and space, for an infinite horizon with a discounted objective and the underlying dynamics driven by a stochastic differential equation. Built upon recent advances in the continuous approach to RL, we develop a notion of occupation time (specifically for a discounted objective), and show how it can be effectively used to derive performance-difference and local-approximation formulas. We further extend these results to illustrate their applications in the PG (policy gradient) and TRPO/PPO (trust region policy optimization/ proximal policy optimization) methods, which have been familiar and powerful tools in the discrete RL setting but under-developed in continuous RL. Through numerical experiments, we demonstrate the effectiveness and advantages of our approach.
♻ ☆ Goal-Conditioned Imitation Learning using Score-based Diffusion Policies
We propose a new policy representation based on score-based diffusion models (SDMs). We apply our new policy representation in the domain of Goal-Conditioned Imitation Learning (GCIL) to learn general-purpose goal-specified policies from large uncurated datasets without rewards. Our new goal-conditioned policy architecture "$\textbf{BE}$havior generation with $\textbf{S}$c$\textbf{O}$re-based Diffusion Policies" (BESO) leverages a generative, score-based diffusion model as its policy. BESO decouples the learning of the score model from the inference sampling process, and, hence allows for fast sampling strategies to generate goal-specified behavior in just 3 denoising steps, compared to 30+ steps of other diffusion based policies. Furthermore, BESO is highly expressive and can effectively capture multi-modality present in the solution space of the play data. Unlike previous methods such as Latent Plans or C-Bet, BESO does not rely on complex hierarchical policies or additional clustering for effective goal-conditioned behavior learning. Finally, we show how BESO can even be used to learn a goal-independent policy from play-data using classifier-free guidance. To the best of our knowledge this is the first work that a) represents a behavior policy based on such a decoupled SDM b) learns an SDM based policy in the domain of GCIL and c) provides a way to simultaneously learn a goal-dependent and a goal-independent policy from play-data. We evaluate BESO through detailed simulation and show that it consistently outperforms several state-of-the-art goal-conditioned imitation learning methods on challenging benchmarks. We additionally provide extensive ablation studies and experiments to demonstrate the effectiveness of our method for goal-conditioned behavior generation. Demonstrations and Code are available at https://intuitive-robots.github.io/beso-website/
comment: Accepted at RSS 2023
♻ ☆ Automatic Photo Orientation Detection with Convolutional Neural Networks
We apply convolutional neural networks (CNN) to the problem of image orientation detection in the context of determining the correct orientation (from 0, 90, 180, and 270 degrees) of a consumer photo. The problem is especially important for digitazing analog photographs. We substantially improve on the published state of the art in terms of the performance on one of the standard datasets, and test our system on a more difficult large dataset of consumer photos. We use Guided Backpropagation to obtain insights into how our CNN detects photo orientation, and to explain its mistakes.
♻ ☆ A Theoretical Analysis of the Learning Dynamics under Class Imbalance
Data imbalance is a common problem in machine learning that can have a critical effect on the performance of a model. Various solutions exist but their impact on the convergence of the learning dynamics is not understood. Here, we elucidate the significant negative impact of data imbalance on learning, showing that the learning curves for minority and majority classes follow sub-optimal trajectories when training with a gradient-based optimizer. This slowdown is related to the imbalance ratio and can be traced back to a competition between the optimization of different classes. Our main contribution is the analysis of the convergence of full-batch (GD) and stochastic gradient descent (SGD), and of variants that renormalize the contribution of each per-class gradient. We find that GD is not guaranteed to decrease the loss for each class but that this problem can be addressed by performing a per-class normalization of the gradient. With SGD, class imbalance has an additional effect on the direction of the gradients: the minority class suffers from a higher directional noise, which reduces the effectiveness of the per-class gradient normalization. Our findings not only allow us to understand the potential and limitations of strategies involving the per-class gradients, but also the reason for the effectiveness of previously used solutions for class imbalance such as oversampling.
comment: changed the structures of the sections to improve the readability, added a section in appendix for multi-class extention
♻ ☆ On Preemption and Learning in Stochastic Scheduling ICML 2023
We study single-machine scheduling of jobs, each belonging to a job type that determines its duration distribution. We start by analyzing the scenario where the type characteristics are known and then move to two learning scenarios where the types are unknown: non-preemptive problems, where each started job must be completed before moving to another job; and preemptive problems, where job execution can be paused in the favor of moving to a different job. In both cases, we design algorithms that achieve sublinear excess cost, compared to the performance with known types, and prove lower bounds for the non-preemptive case. Notably, we demonstrate, both theoretically and through simulations, how preemptive algorithms can greatly outperform non-preemptive ones when the durations of different job types are far from one another, a phenomenon that does not occur when the type durations are known.
comment: Accepted to ICML 2023
♻ ☆ AirfRANS: High Fidelity Computational Fluid Dynamics Dataset for Approximating Reynolds-Averaged Navier-Stokes Solutions
Surrogate models are necessary to optimize meaningful quantities in physical dynamics as their recursive numerical resolutions are often prohibitively expensive. It is mainly the case for fluid dynamics and the resolution of Navier-Stokes equations. However, despite the fast-growing field of data-driven models for physical systems, reference datasets representing real-world phenomena are lacking. In this work, we develop AirfRANS, a dataset for studying the two-dimensional incompressible steady-state Reynolds-Averaged Navier-Stokes equations over airfoils at a subsonic regime and for different angles of attacks. We also introduce metrics on the stress forces at the surface of geometries and visualization of boundary layers to assess the capabilities of models to accurately predict the meaningful information of the problem. Finally, we propose deep learning baselines on four machine learning tasks to study AirfRANS under different constraints for generalization considerations: big and scarce data regime, Reynolds number, and angle of attack extrapolation.
♻ ☆ SQuId: Measuring Speech Naturalness in Many Languages ICASSP 2023
Much of text-to-speech research relies on human evaluation, which incurs heavy costs and slows down the development process. The problem is particularly acute in heavily multilingual applications, where recruiting and polling judges can take weeks. We introduce SQuId (Speech Quality Identification), a multilingual naturalness prediction model trained on over a million ratings and tested in 65 locales-the largest effort of this type to date. The main insight is that training one model on many locales consistently outperforms mono-locale baselines. We present our task, the model, and show that it outperforms a competitive baseline based on w2v-BERT and VoiceMOS by 50.0%. We then demonstrate the effectiveness of cross-locale transfer during fine-tuning and highlight its effect on zero-shot locales, i.e., locales for which there is no fine-tuning data. Through a series of analyses, we highlight the role of non-linguistic effects such as sound artifacts in cross-locale transfer. Finally, we present the effect of our design decision, e.g., model size, pre-training diversity, and language rebalancing with several ablation experiments.
comment: Accepted at ICASSP 2023, with additional material in the appendix
♻ ☆ Provably Learning Diverse Features in Multi-View Data with Midpoint Mixup ICML 2023
Mixup is a data augmentation technique that relies on training using random convex combinations of data points and their labels. In recent years, Mixup has become a standard primitive used in the training of state-of-the-art image classification models due to its demonstrated benefits over empirical risk minimization with regards to generalization and robustness. In this work, we try to explain some of this success from a feature learning perspective. We focus our attention on classification problems in which each class may have multiple associated features (or views) that can be used to predict the class correctly. Our main theoretical results demonstrate that, for a non-trivial class of data distributions with two features per class, training a 2-layer convolutional network using empirical risk minimization can lead to learning only one feature for almost all classes while training with a specific instantiation of Mixup succeeds in learning both features for every class. We also show empirically that these theoretical insights extend to the practical settings of image benchmarks modified to have multiple features.
comment: 37 pages, 2 figures, ICML 2023
♻ ☆ Hiding Data Helps: On the Benefits of Masking for Sparse Coding ICML 2023
Sparse coding, which refers to modeling a signal as sparse linear combinations of the elements of a learned dictionary, has proven to be a successful (and interpretable) approach in applications such as signal processing, computer vision, and medical imaging. While this success has spurred much work on provable guarantees for dictionary recovery when the learned dictionary is the same size as the ground-truth dictionary, work on the setting where the learned dictionary is larger (or over-realized) with respect to the ground truth is comparatively nascent. Existing theoretical results in this setting have been constrained to the case of noise-less data. We show in this work that, in the presence of noise, minimizing the standard dictionary learning objective can fail to recover the elements of the ground-truth dictionary in the over-realized regime, regardless of the magnitude of the signal in the data-generating process. Furthermore, drawing from the growing body of work on self-supervised learning, we propose a novel masking objective for which recovering the ground-truth dictionary is in fact optimal as the signal increases for a large class of data-generating processes. We corroborate our theoretical results with experiments across several parameter regimes showing that our proposed objective also enjoys better empirical performance than the standard reconstruction objective.
comment: 16 pages, 1 figure, ICML 2023
♻ ☆ Beam Tree Recursive Cells ICML 2023
We propose Beam Tree Recursive Cell (BT-Cell) - a backpropagation-friendly framework to extend Recursive Neural Networks (RvNNs) with beam search for latent structure induction. We further extend this framework by proposing a relaxation of the hard top-k operators in beam search for better propagation of gradient signals. We evaluate our proposed models in different out-of-distribution splits in both synthetic and realistic data. Our experiments show that BTCell achieves near-perfect performance on several challenging structure-sensitive synthetic tasks like ListOps and logical inference while maintaining comparable performance in realistic data against other RvNN-based models. Additionally, we identify a previously unknown failure case for neural models in generalization to unseen number of arguments in ListOps. The code is available at: https://github.com/JRC1995/BeamTreeRecursiveCells.
comment: Accepted in ICML 2023
♻ ☆ Some Supervision Required: Incorporating Oracle Policies in Reinforcement Learning via Epistemic Uncertainty Metrics
An inherent problem of reinforcement learning is performing exploration of an environment through random actions, of which a large portion can be unproductive. Instead, exploration can be improved by initializing the learning policy with an existing (previously learned or hard-coded) oracle policy, offline data, or demonstrations. In the case of using an oracle policy, it can be unclear how best to incorporate the oracle policy's experience into the learning policy in a way that maximizes learning sample efficiency. In this paper, we propose a method termed Critic Confidence Guided Exploration (CCGE) for incorporating such an oracle policy into standard actor-critic reinforcement learning algorithms. More specifically, CCGE takes in the oracle policy's actions as suggestions and incorporates this information into the learning scheme when uncertainty is high, while ignoring it when the uncertainty is low. CCGE is agnostic to methods of estimating uncertainty, and we show that it is equally effective with two different techniques. Empirically, we evaluate the effect of CCGE on various benchmark reinforcement learning tasks, and show that this idea can lead to improved sample efficiency and final performance. Furthermore, when evaluated on sparse reward environments, CCGE is able to perform competitively against adjacent algorithms that also leverage an oracle policy. Our experiments show that it is possible to utilize uncertainty as a heuristic to guide exploration using an oracle in reinforcement learning. We expect that this will inspire more research in this direction, where various heuristics are used to determine the direction of guidance provided to learning.
comment: Under review at TMLR
♻ ☆ Finding Materialized Models for Model Reuse
Materialized model query aims to find the most appropriate materialized model as the initial model for model reuse. It is the precondition of model reuse, and has recently attracted much attention. {Nonetheless, the existing methods suffer from the need to provide source data, limited range of applications, and inefficiency since they do not construct a suitable metric to measure the target-related knowledge of materialized models. To address this, we present \textsf{MMQ}, a source-data free, general, efficient, and effective materialized model query framework.} It uses a Gaussian mixture-based metric called separation degree to rank materialized models. For each materialized model, \textsf{MMQ} first vectorizes the samples in the target dataset into probability vectors by directly applying this model, then utilizes Gaussian distribution to fit for each class of probability vectors, and finally uses separation degree on the Gaussian distributions to measure the target-related knowledge of the materialized model. Moreover, we propose an improved \textsf{MMQ} (\textsf{I-MMQ}), which significantly reduces the query time while retaining the query performance of \textsf{MMQ}. Extensive experiments on a range of practical model reuse workloads demonstrate the effectiveness and efficiency of \textsf{MMQ}.
♻ ☆ Scalable Optimal Margin Distribution Machine
Optimal margin Distribution Machine (ODM) is a newly proposed statistical learning framework rooting in the novel margin theory, which demonstrates better generalization performance than the traditional large margin based counterparts. Nonetheless, it suffers from the ubiquitous scalability problem regarding both computation time and memory as other kernel methods. This paper proposes a scalable ODM, which can achieve nearly ten times speedup compared to the original ODM training method. For nonlinear kernels, we propose a novel distribution-aware partition method to make the local ODM trained on each partition be close and converge fast to the global one. When linear kernel is applied, we extend a communication efficient SVRG method to accelerate the training further. Extensive empirical studies validate that our proposed method is highly computational efficient and almost never worsen the generalization.
The SSL Interplay: Augmentations, Inductive Bias, and Generalization
Self-supervised learning (SSL) has emerged as a powerful framework to learn representations from raw data without supervision. Yet in practice, engineers face issues such as instability in tuning optimizers and collapse of representations during training. Such challenges motivate the need for a theory to shed light on the complex interplay between the choice of data augmentation, network architecture, and training algorithm. We study such an interplay with a precise analysis of generalization performance on both pretraining and downstream tasks in a theory friendly setup, and highlight several insights for SSL practitioners that arise from our theory.
♻ ☆ An Advantage Using Feature Selection with a Quantum Annealer
Feature selection is a technique in statistical prediction modeling that identifies features in a record with a strong statistical connection to the target variable. Excluding features with a weak statistical connection to the target variable in training not only drops the dimension of the data, which decreases the time complexity of the algorithm, it also decreases noise within the data which assists in avoiding overfitting. In all, feature selection assists in training a robust statistical model that performs well and is stable. Given the lack of scalability in classical computation, current techniques only consider the predictive power of the feature and not redundancy between the features themselves. Recent advancements in feature selection that leverages quantum annealing (QA) gives a scalable technique that aims to maximize the predictive power of the features while minimizing redundancy. As a consequence, it is expected that this algorithm would assist in the bias/variance trade-off yielding better features for training a statistical model. This paper tests this intuition against classical methods by utilizing open-source data sets and evaluate the efficacy of each trained statistical model well-known prediction algorithms. The numerical results display an advantage utilizing the features selected from the algorithm that leveraged QA.
♻ ☆ Neural Task Synthesis for Visual Programming
Generative neural models hold great promise in enhancing programming education by synthesizing new content for students. We seek to design neural models that can automatically generate programming tasks for a given specification in the context of visual programming domains. Despite the recent successes of large generative models like GPT-4, our initial results show that these models are ineffective in synthesizing visual programming tasks and struggle with logical and spatial reasoning. We propose a novel neuro-symbolic technique, NeurTaskSyn, that can synthesize programming tasks for a specification given in the form of desired programming concepts exercised by its solution code and constraints on the visual task. NeurTaskSyn has two components: the first component is trained via imitation learning procedure to generate possible solution codes, and the second component is trained via reinforcement learning procedure to guide an underlying symbolic execution engine that generates visual tasks for these codes. We demonstrate the effectiveness of NeurTaskSyn through an extensive empirical evaluation and a qualitative study on reference tasks taken from the Hour of Code: Classic Maze challenge by Code-dot-org and the Intro to Programming with Karel course by CodeHS-dot-com.
♻ ☆ A Survey on Causal Reinforcement Learning
While Reinforcement Learning (RL) achieves tremendous success in sequential decision-making problems of many domains, it still faces key challenges of data inefficiency and the lack of interpretability. Interestingly, many researchers have leveraged insights from the causality literature recently, bringing forth flourishing works to unify the merits of causality and address well the challenges from RL. As such, it is of great necessity and significance to collate these Causal Reinforcement Learning (CRL) works, offer a review of CRL methods, and investigate the potential functionality from causality toward RL. In particular, we divide existing CRL approaches into two categories according to whether their causality-based information is given in advance or not. We further analyze each category in terms of the formalization of different models, ranging from the Markov Decision Process (MDP), Partially Observed Markov Decision Process (POMDP), Multi-Arm Bandits (MAB), and Dynamic Treatment Regime (DTR). Moreover, we summarize the evaluation matrices and open sources while we discuss emerging applications, along with promising prospects for the future development of CRL.
comment: 29 pages, 20 figures
♻ ☆ A Meta-learning Framework for Tuning Parameters of Protection Mechanisms in Trustworthy Federated Learning
Trustworthy Federated Learning (TFL) typically leverages protection mechanisms to guarantee privacy. However, protection mechanisms inevitably introduce utility loss or efficiency reduction while protecting data privacy. Therefore, protection mechanisms and their parameters should be carefully chosen to strike an optimal tradeoff between \textit{privacy leakage}, \textit{utility loss}, and \textit{efficiency reduction}. To this end, federated learning practitioners need tools to measure the three factors and optimize the tradeoff between them to choose the protection mechanism that is most appropriate to the application at hand. Motivated by this requirement, we propose a framework that (1) formulates TFL as a problem of finding a protection mechanism to optimize the tradeoff between privacy leakage, utility loss, and efficiency reduction and (2) formally defines bounded measurements of the three factors. We then propose a meta-learning algorithm to approximate this optimization problem and find optimal protection parameters for representative protection mechanisms, including Randomization, Homomorphic Encryption, Secret Sharing, and Compression. We further design estimation algorithms to quantify these found optimal protection parameters in a practical horizontal federated learning setting and provide a theoretical analysis of the estimation error.
comment: arXiv admin note: text overlap with arXiv:2209.00230
♻ ☆ Nonlinear Advantage: Trained Networks Might Not Be As Complex as You Think
We perform an empirical study of the behaviour of deep networks when fully linearizing some of its feature channels through a sparsity prior on the overall number of nonlinear units in the network. In experiments on image classification and machine translation tasks, we investigate how much we can simplify the network function towards linearity before performance collapses. First, we observe a significant performance gap when reducing nonlinearity in the network function early on as opposed to late in training, in-line with recent observations on the time-evolution of the data-dependent NTK. Second, we find that after training, we are able to linearize a significant number of nonlinear units while maintaining a high performance, indicating that much of a network's expressivity remains unused but helps gradient descent in early stages of training. To characterize the depth of the resulting partially linearized network, we introduce a measure called average path length, representing the average number of active nonlinearities encountered along a path in the network graph. Under sparsity pressure, we find that the remaining nonlinear units organize into distinct structures, forming core-networks of near constant effective depth and width, which in turn depend on task difficulty.
♻ ☆ Mechanistic Mode Connectivity ICML
We study neural network loss landscapes through the lens of mode connectivity, the observation that minimizers of neural networks retrieved via training on a dataset are connected via simple paths of low loss. Specifically, we ask the following question: are minimizers that rely on different mechanisms for making their predictions connected via simple paths of low loss? We provide a definition of mechanistic similarity as shared invariances to input transformations and demonstrate that lack of linear connectivity between two models implies they use dissimilar mechanisms for making their predictions. Relevant to practice, this result helps us demonstrate that naive fine-tuning on a downstream dataset can fail to alter a model's mechanisms, e.g., fine-tuning can fail to eliminate a model's reliance on spurious attributes. Our analysis also motivates a method for targeted alteration of a model's mechanisms, named connectivity-based fine-tuning (CBFT), which we analyze using several synthetic datasets for the task of reducing a model's reliance on spurious attributes.
comment: ICML, 2023
♻ ☆ InGram: Inductive Knowledge Graph Embedding via Relation Graphs ICML 2023
Inductive knowledge graph completion has been considered as the task of predicting missing triplets between new entities that are not observed during training. While most inductive knowledge graph completion methods assume that all entities can be new, they do not allow new relations to appear at inference time. This restriction prohibits the existing methods from appropriately handling real-world knowledge graphs where new entities accompany new relations. In this paper, we propose an INductive knowledge GRAph eMbedding method, InGram, that can generate embeddings of new relations as well as new entities at inference time. Given a knowledge graph, we define a relation graph as a weighted graph consisting of relations and the affinity weights between them. Based on the relation graph and the original knowledge graph, InGram learns how to aggregate neighboring embeddings to generate relation and entity embeddings using an attention mechanism. Experimental results show that InGram outperforms 14 different state-of-the-art methods on varied inductive learning scenarios.
comment: 14 pages, 4 figures, 6 tables, 40th International Conference on Machine Learning (ICML 2023)
♻ ☆ Label Embedding by Johnson-Lindenstrauss Matrices
We present a simple and scalable framework for extreme multiclass classification based on Johnson-Lindenstrauss matrices (JLMs). Using the columns of a JLM to embed the labels, a $C$-class classification problem is transformed into a regression problem with $\cO(\log C)$ output dimension. We derive an excess risk bound, revealing a tradeoff between computational efficiency and prediction accuracy, and further show that under the Massart noise condition, the penalty for dimension reduction vanishes. Our approach is easily parallelizable, and experimental results demonstrate its effectiveness and scalability in large-scale applications.
♻ ☆ Resolution Limits of Non-Adaptive 20 Questions Search for a Moving Target
Using the 20 questions estimation framework with query-dependent noise, we study non-adaptive search strategies for a moving target over the unit cube with unknown initial location and velocities under a piecewise constant velocity model. In this search problem, there is an oracle who knows the instantaneous location of the target at any time. Our task is to query the oracle as few times as possible to accurately estimate the location of the target at any specified time. We first study the case where the oracle's answer to each query is corrupted by discrete noise and then generalize our results to the case of additive white Gaussian noise. In our formulation, the performance criterion is the resolution, which is defined as the maximal $L_\infty$ distance between the true locations and estimated locations. We characterize the minimal resolution of an optimal non-adaptive query procedure with a finite number of queries by deriving non-asymptotic and asymptotic bounds. Our bounds are tight in the first-order asymptotic sense when the number of queries satisfies a certain condition and our bounds are tight in the stronger second-order asymptotic sense when the target moves with a constant velocity. To prove our results, we relate the current problem to channel coding, borrow ideas from finite blocklength information theory and construct bounds on the number of possible quantized target trajectories.
comment: To appear in IEEE Transactions on Information Theory
♻ ☆ Toward Foundation Models for Earth Monitoring: Generalizable Deep Learning Models for Natural Hazard Segmentation
Climate change results in an increased probability of extreme weather events that put societies and businesses at risk on a global scale. Therefore, near real-time mapping of natural hazards is an emerging priority for the support of natural disaster relief, risk management, and informing governmental policy decisions. Recent methods to achieve near real-time mapping increasingly leverage deep learning (DL). However, DL-based approaches are designed for one specific task in a single geographic region based on specific frequency bands of satellite data. Therefore, DL models used to map specific natural hazards struggle with their generalization to other types of natural hazards in unseen regions. In this work, we propose a methodology to significantly improve the generalizability of DL natural hazards mappers based on pre-training on a suitable pre-task. Without access to any data from the target domain, we demonstrate this improved generalizability across four U-Net architectures for the segmentation of unseen natural hazards. Importantly, our method is invariant to geographic differences and differences in the type of frequency bands of satellite data. By leveraging characteristics of unlabeled images from the target domain that are publicly available, our approach is able to further improve the generalization behavior without fine-tuning. Thereby, our approach supports the development of foundation models for earth monitoring with the objective of directly segmenting unseen natural hazards across novel geographic regions given different sources of satellite imagery.
comment: Accepted at IEEE International Geoscience and Remote Sensing Symposium (IGARSS 2023)
♻ ☆ Critical Points and Convergence Analysis of Generative Deep Linear Networks Trained with Bures-Wasserstein Loss ICML 2023
We consider a deep matrix factorization model of covariance matrices trained with the Bures-Wasserstein distance. While recent works have made important advances in the study of the optimization problem for overparametrized low-rank matrix approximation, much emphasis has been placed on discriminative settings and the square loss. In contrast, our model considers another interesting type of loss and connects with the generative setting. We characterize the critical points and minimizers of the Bures-Wasserstein distance over the space of rank-bounded matrices. For low-rank matrices the Hessian of this loss can theoretically blow up, which creates challenges to analyze convergence of optimizaton methods. We establish convergence results for gradient flow using a smooth perturbative version of the loss and convergence results for finite step size gradient descent under certain assumptions on the initial weights.
comment: 35 pages, 1 figure, accepted at ICML 2023
♻ ☆ K-SHAP: Policy Clustering Algorithm for Anonymous State-Action Pairs ICML 2023
Learning agent behaviors from observational data has shown to improve our understanding of their decision-making processes, advancing our ability to explain their interactions with the environment and other agents. While multiple learning techniques have been proposed in the literature, there is one particular setting that has not been explored yet: multi agent systems where agent identities remain anonymous. For instance, in financial markets labeled data that identifies market participant strategies is typically proprietary, and only the anonymous state-action pairs that result from the interaction of multiple market participants are publicly available. As a result, sequences of agent actions are not observable, restricting the applicability of existing work. In this paper, we propose a Policy Clustering algorithm, called K-SHAP, that learns to group anonymous state-action pairs according to the agent policies. We frame the problem as an Imitation Learning (IL) task, and we learn a world-policy able to mimic all the agent behaviors upon different environmental states. We leverage the world-policy to explain each anonymous observation through an additive feature attribution method called SHAP (SHapley Additive exPlanations). Finally, by clustering the explanations we show that we are able to identify different agent policies and group observations accordingly. We evaluate our approach on simulated synthetic market data and a real-world financial dataset. We show that our proposal significantly and consistently outperforms the existing methods, identifying different agent strategies.
comment: ICML 2023
♻ ☆ Canary in a Coalmine: Better Membership Inference with Ensembled Adversarial Queries ICLR 2023
As industrial applications are increasingly automated by machine learning models, enforcing personal data ownership and intellectual property rights requires tracing training data back to their rightful owners. Membership inference algorithms approach this problem by using statistical techniques to discern whether a target sample was included in a model's training set. However, existing methods only utilize the unaltered target sample or simple augmentations of the target to compute statistics. Such a sparse sampling of the model's behavior carries little information, leading to poor inference capabilities. In this work, we use adversarial tools to directly optimize for queries that are discriminative and diverse. Our improvements achieve significantly more accurate membership inference than existing methods, especially in offline scenarios and in the low false-positive regime which is critical in legal settings. Code is available at https://github.com/YuxinWenRick/canary-in-a-coalmine.
comment: Code is available at https://github.com/YuxinWenRick/canary-in-a-coalmine, published as a conference paper at ICLR 2023
♻ ☆ How to Attain Communication-Efficient DNN Training? Convert, Compress, Correct
This paper introduces CO3 -- an algorithm for communication-efficient federated Deep Neural Network (DNN) training. CO3 takes its name from three processing applied which reduce the communication load when transmitting the local DNN gradients from the remote users to the Parameter Server. Namely: (i) gradient quantization through floating-point conversion, (ii) lossless compression of the quantized gradient, and (iii) quantization error correction. We carefully design each of the steps above to assure good training performance under a constraint on the communication rate. In particular, in steps (i) and (ii), we adopt the assumption that DNN gradients are distributed according to a generalized normal distribution, which is validated numerically in the paper. For step (iii), we utilize an error feedback with memory decay mechanism to correct the quantization error introduced in step (i). We argue that the memory decay coefficient, similarly to the learning rate, can be optimally tuned to improve convergence. A rigorous convergence analysis of the proposed CO3 with SGD is provided. Moreover, with extensive simulations, we show that CO3 offers improved performance when compared with existing gradient compression schemes in the literature which employ sketching and non-uniform quantization of the local gradients.
comment: arXiv admin note: substantial text overlap with arXiv:2203.09044
♻ ☆ Hard Prompts Made Easy: Gradient-Based Discrete Optimization for Prompt Tuning and Discovery
The strength of modern generative models lies in their ability to be controlled through text-based prompts. Typical "hard" prompts are made from interpretable words and tokens, and must be hand-crafted by humans. There are also "soft" prompts, which consist of continuous feature vectors. These can be discovered using powerful optimization methods, but they cannot be easily interpreted, re-used across models, or plugged into a text-based interface. We describe an approach to robustly optimize hard text prompts through efficient gradient-based optimization. Our approach automatically generates hard text-based prompts for both text-to-image and text-to-text applications. In the text-to-image setting, the method creates hard prompts for diffusion models, allowing API users to easily generate, discover, and mix and match image concepts without prior knowledge on how to prompt the model. In the text-to-text setting, we show that hard prompts can be automatically discovered that are effective in tuning LMs for classification.
comment: 15 pages, 12 figures, Code is available at https://github.com/YuxinWenRick/hard-prompts-made-easy
♻ ☆ Best Arm Identification for Stochastic Rising Bandits
Stochastic Rising Bandits (SRBs) model sequential decision-making problems in which the expected rewards of the available options increase every time they are selected. This setting captures a wide range of scenarios in which the available options are learning entities whose performance improves (in expectation) over time. While previous works addressed the regret minimization problem, this paper, focuses on the fixed-budget Best Arm Identification (BAI) problem for SRBs. In this scenario, given a fixed budget of rounds, we are asked to provide a recommendation about the best option at the end of the identification process. We propose two algorithms to tackle the above-mentioned setting, namely R-UCBE, which resorts to a UCB-like approach, and R-SR, which employs a successive reject procedure. Then, we prove that, with a sufficiently large budget, they provide guarantees on the probability of properly identifying the optimal option at the end of the learning process. Furthermore, we derive a lower bound on the error probability, matched by our R-SR (up to logarithmic factors), and illustrate how the need for a sufficiently large budget is unavoidable in the SRB setting. Finally, we numerically validate the proposed algorithms in both synthetic and real-world environments and compare them with the currently available BAI strategies.
♻ ☆ Counterfactual Explainer Framework for Deep Reinforcement Learning Models Using Policy Distillation
Deep Reinforcement Learning (DRL) has demonstrated promising capability in solving complex control problems. However, DRL applications in safety-critical systems are hindered by the inherent lack of robust verification techniques to assure their performance in such applications. One of the key requirements of the verification process is the development of effective techniques to explain the system functionality, i.e., why the system produces specific results in given circumstances. Recently, interpretation methods based on the Counterfactual (CF) explanation approach have been proposed to address the problem of explanation in DRLs. This paper proposes a novel CF explanation framework to explain the decisions made by a black-box DRL. To evaluate the efficacy of the proposed explanation framework, we carried out several experiments in the domains of automated driving systems and Atari Pong game. Our analysis demonstrates that the proposed framework generates plausible and meaningful explanations for various decisions made by deep underlying DRLs. Source codes are available at: \url{https://github.com/Amir-Samadi/Counterfactual-Explanation}
♻ ☆ Teaching Small Language Models to Reason
Chain of thought prompting successfully improves the reasoning capabilities of large language models, achieving state of the art results on a range of datasets. However, these reasoning capabilities only appear to emerge in models with a size of over 100 billion parameters. In this paper, we explore the transfer of such reasoning capabilities to models with less than 100 billion parameters via knowledge distillation. Specifically, we finetune a student model on the chain of thought outputs generated by a larger teacher model. Our experiments show that the proposed method improves task performance across arithmetic, commonsense and symbolic reasoning datasets. For example, the accuracy of T5 XXL on GSM8K improves from 8.11% to 21.99% when finetuned on PaLM-540B generated chains of thought.
♻ ☆ Representation Learning on Hyper-Relational and Numeric Knowledge Graphs with Transformers KDD
A hyper-relational knowledge graph has been recently studied where a triplet is associated with a set of qualifiers; a qualifier is composed of a relation and an entity, providing auxiliary information for a triplet. While existing hyper-relational knowledge graph embedding methods assume that the entities are discrete objects, some information should be represented using numeric values, e.g., (J.R.R., was born in, 1892). Also, a triplet (J.R.R., educated at, Oxford Univ.) can be associated with a qualifier such as (start time, 1911). In this paper, we propose a unified framework named HyNT that learns representations of a hyper-relational knowledge graph containing numeric literals in either triplets or qualifiers. We define a context transformer and a prediction transformer to learn the representations based not only on the correlations between a triplet and its qualifiers but also on the numeric information. By learning compact representations of triplets and qualifiers and feeding them into the transformers, we reduce the computation cost of using transformers. Using HyNT, we can predict missing numeric values in addition to missing entities or relations in a hyper-relational knowledge graph. Experimental results show that HyNT significantly outperforms state-of-the-art methods on real-world datasets.
comment: 11 pages, 5 figures, 12 tables. 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2023)
♻ ☆ Neural Architecture Search for Energy Efficient Always-on Audio Models
Mobile and edge computing devices for always-on classification tasks require energy-efficient neural network architectures. In this paper we present several changes to neural architecture searches (NAS) that improve the chance of success in practical situations. Our search simultaneously optimizes for network accuracy, energy efficiency and memory usage. We benchmark the performance of our search on real hardware, but since running thousands of tests with real hardware is difficult we use a random forest model to roughly predict the energy usage of a candidate network. We present a search strategy that uses both Bayesian and regularized evolutionary search with particle swarms, and employs early-stopping to reduce the computational burden. Our search, evaluated on a sound-event classification dataset based upon AudioSet, results in an order of magnitude less energy per inference and a much smaller memory footprint than our baseline MobileNetV1/V2 implementations while slightly improving task accuracy. We also demonstrate how combining a 2D spectrogram with a convolution with many filters causes a computational bottleneck for audio classification and that alternative approaches reduce the computational burden but sacrifice task accuracy.
♻ ☆ Tree-Ring Watermarks: Fingerprints for Diffusion Images that are Invisible and Robust
Watermarking the outputs of generative models is a crucial technique for tracing copyright and preventing potential harm from AI-generated content. In this paper, we introduce a novel technique called Tree-Ring Watermarking that robustly fingerprints diffusion model outputs. Unlike existing methods that perform post-hoc modifications to images after sampling, Tree-Ring Watermarking subtly influences the entire sampling process, resulting in a model fingerprint that is invisible to humans. The watermark embeds a pattern into the initial noise vector used for sampling. These patterns are structured in Fourier space so that they are invariant to convolutions, crops, dilations, flips, and rotations. After image generation, the watermark signal is detected by inverting the diffusion process to retrieve the noise vector, which is then checked for the embedded signal. We demonstrate that this technique can be easily applied to arbitrary diffusion models, including text-conditioned Stable Diffusion, as a plug-in with negligible loss in FID. Our watermark is semantically hidden in the image space and is far more robust than watermarking alternatives that are currently deployed. Code is available at https://github.com/YuxinWenRick/tree-ring-watermark.
comment: 16 pages, 8 figures, code is available at https://github.com/YuxinWenRick/tree-ring-watermark, fixed the repo link
♻ ☆ Coin Sampling: Gradient-Based Bayesian Inference without Learning Rates ICML 2023
In recent years, particle-based variational inference (ParVI) methods such as Stein variational gradient descent (SVGD) have grown in popularity as scalable methods for Bayesian inference. Unfortunately, the properties of such methods invariably depend on hyperparameters such as the learning rate, which must be carefully tuned by the practitioner in order to ensure convergence to the target measure at a suitable rate. In this paper, we introduce a suite of new particle-based methods for scalable Bayesian inference based on coin betting, which are entirely learning-rate free. We illustrate the performance of our approach on a range of numerical examples, including several high-dimensional models and datasets, demonstrating comparable performance to other ParVI algorithms with no need to tune a learning rate.
comment: ICML 2023
♻ ☆ Structural Optimization of Factor Graphs for Symbol Detection via Continuous Clustering and Machine Learning ICASSP 2023
We propose a novel method to optimize the structure of factor graphs for graph-based inference. As an example inference task, we consider symbol detection on linear inter-symbol interference channels. The factor graph framework has the potential to yield low-complexity symbol detectors. However, the sum-product algorithm on cyclic factor graphs is suboptimal and its performance is highly sensitive to the underlying graph. Therefore, we optimize the structure of the underlying factor graphs in an end-to-end manner using machine learning. For that purpose, we transform the structural optimization into a clustering problem of low-degree factor nodes that incorporates the known channel model into the optimization. Furthermore, we study the combination of this approach with neural belief propagation, yielding near-maximum a posteriori symbol detection performance for specific channels.
comment: Accepted at ICASSP 2023
Multimedia 4
☆ 4DSR-GCN: 4D Video Point Cloud Upsampling using Graph Convolutional Networks
Time varying sequences of 3D point clouds, or 4D point clouds, are now being acquired at an increasing pace in several applications (e.g., LiDAR in autonomous or assisted driving). In many cases, such volume of data is transmitted, thus requiring that proper compression tools are applied to either reduce the resolution or the bandwidth. In this paper, we propose a new solution for upscaling and restoration of time-varying 3D video point clouds after they have been heavily compressed. In consideration of recent growing relevance of 3D applications, %We focused on a model allowing user-side upscaling and artifact removal for 3D video point clouds, a real-time stream of which would require . Our model consists of a specifically designed Graph Convolutional Network (GCN) that combines Dynamic Edge Convolution and Graph Attention Networks for feature aggregation in a Generative Adversarial setting. By taking inspiration PointNet++, We present a different way to sample dense point clouds with the intent to make these modules work in synergy to provide each node enough features about its neighbourhood in order to later on generate new vertices. Compared to other solutions in the literature that address the same task, our proposed model is capable of obtaining comparable results in terms of quality of the reconstruction, while using a substantially lower number of parameters (about 300KB), making our solution deployable in edge computing devices such as LiDAR.
☆ PV2TEA: Patching Visual Modality to Textual-Established Information Extraction ACL 2023
Information extraction, e.g., attribute value extraction, has been extensively studied and formulated based only on text. However, many attributes can benefit from image-based extraction, like color, shape, pattern, among others. The visual modality has long been underutilized, mainly due to multimodal annotation difficulty. In this paper, we aim to patch the visual modality to the textual-established attribute information extractor. The cross-modality integration faces several unique challenges: (C1) images and textual descriptions are loosely paired intra-sample and inter-samples; (C2) images usually contain rich backgrounds that can mislead the prediction; (C3) weakly supervised labels from textual-established extractors are biased for multimodal training. We present PV2TEA, an encoder-decoder architecture equipped with three bias reduction schemes: (S1) Augmented label-smoothed contrast to improve the cross-modality alignment for loosely-paired image and text; (S2) Attention-pruning that adaptively distinguishes the visual foreground; (S3) Two-level neighborhood regularization that mitigates the label textual bias via reliability estimation. Empirical results on real-world e-Commerce datasets demonstrate up to 11.74% absolute (20.97% relatively) F1 increase over unimodal baselines.
comment: ACL 2023 Findings
♻ ☆ GPA-Net:No-Reference Point Cloud Quality Assessment with Multi-task Graph Convolutional Network
With the rapid development of 3D vision, point cloud has become an increasingly popular 3D visual media content. Due to the irregular structure, point cloud has posed novel challenges to the related research, such as compression, transmission, rendering and quality assessment. In these latest researches, point cloud quality assessment (PCQA) has attracted wide attention due to its significant role in guiding practical applications, especially in many cases where the reference point cloud is unavailable. However, current no-reference metrics which based on prevalent deep neural network have apparent disadvantages. For example, to adapt to the irregular structure of point cloud, they require preprocessing such as voxelization and projection that introduce extra distortions, and the applied grid-kernel networks, such as Convolutional Neural Networks, fail to extract effective distortion-related features. Besides, they rarely consider the various distortion patterns and the philosophy that PCQA should exhibit shifting, scaling, and rotational invariance. In this paper, we propose a novel no-reference PCQA metric named the Graph convolutional PCQA network (GPA-Net). To extract effective features for PCQA, we propose a new graph convolution kernel, i.e., GPAConv, which attentively captures the perturbation of structure and texture. Then, we propose the multi-task framework consisting of one main task (quality regression) and two auxiliary tasks (distortion type and degree predictions). Finally, we propose a coordinate normalization module to stabilize the results of GPAConv under shift, scale and rotation transformations. Experimental results on two independent databases show that GPA-Net achieves the best performance compared to the state-of-the-art no-reference PCQA metrics, even better than some full-reference metrics in some cases.
♻ ☆ Dance2MIDI: Dance-driven multi-instruments music generation
Dance-driven music generation aims to generate musical pieces conditioned on dance videos. Previous works focus on monophonic or raw audio generation, while the multiinstruments scenario is under-explored. The challenges of the dance-driven multi-instruments music (MIDI) generation are two-fold: 1) no publicly available multi-instruments MIDI and video paired dataset and 2) the weak correlation between music and video. To tackle these challenges, we build the first multi-instruments MIDI and dance paired dataset (D2MIDI). Based on our proposed dataset, we introduce a multi-instruments MIDI generation framework (Dance2MIDI) conditioned on dance video. Specifically, 1) to model the correlation between music and dance, we encode the dance motion using the GCN, and 2) to generate harmonious and coherent music, we employ Transformer to decode the MIDI sequence. We evaluate the generated music of our framework trained on D2MIDI dataset and demonstrate that our method outperforms existing methods. The data and code are available on the GitHub website.
Computation and Language 122
☆ Improving CLIP Training with Language Rewrites
Contrastive Language-Image Pre-training (CLIP) stands as one of the most effective and scalable methods for training transferable vision models using paired image and text data. CLIP models are trained using contrastive loss, which typically relies on data augmentations to prevent overfitting and shortcuts. However, in the CLIP training paradigm, data augmentations are exclusively applied to image inputs, while language inputs remain unchanged throughout the entire training process, limiting the exposure of diverse texts to the same image. In this paper, we introduce Language augmented CLIP (LaCLIP), a simple yet highly effective approach to enhance CLIP training through language rewrites. Leveraging the in-context learning capability of large language models, we rewrite the text descriptions associated with each image. These rewritten texts exhibit diversity in sentence structure and vocabulary while preserving the original key concepts and meanings. During training, LaCLIP randomly selects either the original texts or the rewritten versions as text augmentations for each image. Extensive experiments on CC3M, CC12M, RedCaps and LAION-400M datasets show that CLIP pre-training with language rewrites significantly improves the transfer performance without computation or memory overhead during training. Specifically for ImageNet zero-shot accuracy, LaCLIP outperforms CLIP by 8.2% on CC12M and 2.4% on LAION-400M. Code is available at https://github.com/LijieFan/LaCLIP.
☆ Findings of the VarDial Evaluation Campaign 2023
This report presents the results of the shared tasks organized as part of the VarDial Evaluation Campaign 2023. The campaign is part of the tenth workshop on Natural Language Processing (NLP) for Similar Languages, Varieties and Dialects (VarDial), co-located with EACL 2023. Three separate shared tasks were included this year: Slot and intent detection for low-resource language varieties (SID4LR), Discriminating Between Similar Languages -- True Labels (DSL-TL), and Discriminating Between Similar Languages -- Speech (DSL-S). All three tasks were organized for the first time this year.
☆ Decision-Oriented Dialogue for Human-AI Collaboration
We describe a class of tasks called dialogue decision problems, in which AI assistants must collaborate with one or more humans via natural language to help them make complex decisions. We formalize three domains in which users face everyday decisions: (1) choosing an assignment of reviewers to conference papers, (2) planning a multi-step itinerary in a city, and (3) negotiating travel plans for a group of friends. In each of these settings, AI assistants and users have disparate abilities that they must combine to arrive at the best decision: assistants can access and process large amounts of information, while users have preferences and constraints external to the system. For each task, we build a dialogue environment where agents receive a reward based on the quality of the final decision they reach. Using these environments, we collect human-human dialogues with humans playing the role of assistant. To compare how current AI assistants communicate in these settings, we present baselines using large language models in self-play. Finally, we highlight a number of challenges models face in decision-oriented dialogues, ranging from efficient communication to reasoning and optimization, and release our environments as a testbed for future modeling work.
☆ Let's Verify Step by Step
In recent years, large language models have greatly improved in their ability to perform complex multi-step reasoning. However, even state-of-the-art models still regularly produce logical mistakes. To train more reliable models, we can turn either to outcome supervision, which provides feedback for a final result, or process supervision, which provides feedback for each intermediate reasoning step. Given the importance of training reliable models, and given the high cost of human feedback, it is important to carefully compare the both methods. Recent work has already begun this comparison, but many questions still remain. We conduct our own investigation, finding that process supervision significantly outperforms outcome supervision for training models to solve problems from the challenging MATH dataset. Our process-supervised model solves 78% of problems from a representative subset of the MATH test set. Additionally, we show that active learning significantly improves the efficacy of process supervision. To support related research, we also release PRM800K, the complete dataset of 800,000 step-level human feedback labels used to train our best reward model.
☆ Computational Language Assessment in patients with speech, language, and communication impairments
Speech, language, and communication symptoms enable the early detection, diagnosis, treatment planning, and monitoring of neurocognitive disease progression. Nevertheless, traditional manual neurologic assessment, the speech and language evaluation standard, is time-consuming and resource-intensive for clinicians. We argue that Computational Language Assessment (C.L.A.) is an improvement over conventional manual neurological assessment. Using machine learning, natural language processing, and signal processing, C.L.A. provides a neuro-cognitive evaluation of speech, language, and communication in elderly and high-risk individuals for dementia. ii. facilitates the diagnosis, prognosis, and therapy efficacy in at-risk and language-impaired populations; and iii. allows easier extensibility to assess patients from a wide range of languages. Also, C.L.A. employs Artificial Intelligence models to inform theory on the relationship between language symptoms and their neural bases. It significantly advances our ability to optimize the prevention and treatment of elderly individuals with communication disorders, allowing them to age gracefully with social engagement.
comment: 36 pages, 2 figures, to be submited
☆ ActiveAED: A Human in the Loop Improves Annotation Error Detection ACL 2023
Manually annotated datasets are crucial for training and evaluating Natural Language Processing models. However, recent work has discovered that even widely-used benchmark datasets contain a substantial number of erroneous annotations. This problem has been addressed with Annotation Error Detection (AED) models, which can flag such errors for human re-annotation. However, even though many of these AED methods assume a final curation step in which a human annotator decides whether the annotation is erroneous, they have been developed as static models without any human-in-the-loop component. In this work, we propose ActiveAED, an AED method that can detect errors more accurately by repeatedly querying a human for error corrections in its prediction loop. We evaluate ActiveAED on eight datasets spanning five different tasks and find that it leads to improvements over the state of the art on seven of them, with gains of up to six percentage points in average precision.
comment: Findings of ACL 2023
☆ Monotonic Location Attention for Length Generalization ICML 2023
We explore different ways to utilize position-based cross-attention in seq2seq networks to enable length generalization in algorithmic tasks. We show that a simple approach of interpolating the original and reversed encoded representations combined with relative attention allows near-perfect length generalization for both forward and reverse lookup tasks or copy tasks that had been generally hard to tackle. We also devise harder diagnostic tasks where the relative distance of the ideal attention position varies with timestep. In such settings, the simple interpolation trick with relative attention is not sufficient. We introduce novel variants of location attention building on top of Dubois et al. (2020) to address the new diagnostic tasks. We also show the benefits of our approaches for length generalization in SCAN (Lake & Baroni, 2018) and CFQ (Keysers et al., 2020). Our code is available on GitHub.
comment: Accepted in ICML 2023
☆ Scalable Learning of Latent Language Structure With Logical Offline Cycle Consistency
We introduce Logical Offline Cycle Consistency Optimization (LOCCO), a scalable, semi-supervised method for training a neural semantic parser. Conceptually, LOCCO can be viewed as a form of self-learning where the semantic parser being trained is used to generate annotations for unlabeled text that are then used as new supervision. To increase the quality of annotations, our method utilizes a count-based prior over valid formal meaning representations and a cycle-consistency score produced by a neural text generation model as additional signals. Both the prior and semantic parser are updated in an alternate fashion from full passes over the training data, which can be seen as approximating the marginalization of latent structures through stochastic variational inference. The use of a count-based prior, frozen text generation model, and offline annotation process yields an approach with negligible complexity and latency increases as compared to conventional self-learning. As an added bonus, the annotations produced by LOCCO can be trivially repurposed to train a neural text generation model. We demonstrate the utility of LOCCO on the well-known WebNLG benchmark where we obtain an improvement of 2 points against a self-learning parser under equivalent conditions, an improvement of 1.3 points against the previous state-of-the-art parser, and competitive text generation performance in terms of BLEU score.
☆ Human or Not? A Gamified Approach to the Turing Test
We present "Human or Not?", an online game inspired by the Turing test, that measures the capability of AI chatbots to mimic humans in dialog, and of humans to tell bots from other humans. Over the course of a month, the game was played by over 1.5 million users who engaged in anonymous two-minute chat sessions with either another human or an AI language model which was prompted to behave like humans. The task of the players was to correctly guess whether they spoke to a person or to an AI. This largest scale Turing-style test conducted to date revealed some interesting facts. For example, overall users guessed the identity of their partners correctly in only 68% of the games. In the subset of the games in which users faced an AI bot, users had even lower correct guess rates of 60% (that is, not much higher than chance). This white paper details the development, deployment, and results of this unique experiment. While this experiment calls for many extensions and refinements, these findings already begin to shed light on the inevitable near future which will commingle humans and AI.
comment: 11 pages, 6 figures
☆ Beam Tree Recursive Cells ICML 2023
We propose Beam Tree Recursive Cell (BT-Cell) - a backpropagation-friendly framework to extend Recursive Neural Networks (RvNNs) with beam search for latent structure induction. We further extend this framework by proposing a relaxation of the hard top-k operators in beam search for better propagation of gradient signals. We evaluate our proposed models in different out-of-distribution splits in both synthetic and realistic data. Our experiments show that BTCell achieves near-perfect performance on several challenging structure-sensitive synthetic tasks like ListOps and logical inference while maintaining comparable performance in realistic data against other RvNN-based models. Additionally, we identify a previously unknown failure case for neural models in generalization to unseen number of arguments in ListOps. The code is available at: https://github.com/JRC1995/ BeamTreeRecursiveCells.
comment: Accepted in ICML 2023
☆ Efficient Shapley Values Estimation by Amortization for Text Classification ACL 2023
Despite the popularity of Shapley Values in explaining neural text classification models, computing them is prohibitive for large pretrained models due to a large number of model evaluations. In practice, Shapley Values are often estimated with a small number of stochastic model evaluations. However, we show that the estimated Shapley Values are sensitive to random seed choices -- the top-ranked features often have little overlap across different seeds, especially on examples with longer input texts. This can only be mitigated by aggregating thousands of model evaluations, which on the other hand, induces substantial computational overheads. To mitigate the trade-off between stability and efficiency, we develop an amortized model that directly predicts each input feature's Shapley Value without additional model evaluations. It is trained on a set of examples whose Shapley Values are estimated from a large number of model evaluations to ensure stability. Experimental results on two text classification datasets demonstrate that our amortized model estimates Shapley Values accurately with up to 60 times speedup compared to traditional methods. Furthermore, the estimated values are stable as the inference is deterministic. We release our code at https://github.com/yangalan123/Amortized-Interpretability.
comment: ACL 2023 Camera Ready
☆ MedNgage: A Dataset for Understanding Engagement in Patient-Nurse Conversations ACL
Patients who effectively manage their symptoms often demonstrate higher levels of engagement in conversations and interventions with healthcare practitioners. This engagement is multifaceted, encompassing cognitive and socio-affective dimensions. Consequently, it is crucial for AI systems to understand the engagement in natural conversations between patients and practitioners to better contribute toward patient care. In this paper, we present a novel dataset (MedNgage), which consists of patient-nurse conversations about cancer symptom management. We manually annotate the dataset with a novel framework of categories of patient engagement from two different angles, namely: i) socio-affective engagement (3.1K spans), and ii) cognitive engagement (1.8K spans). Through statistical analysis of the data that is annotated using our framework, we show a positive correlation between patient symptom management outcomes and their engagement in conversations. Additionally, we demonstrate that pre-trained transformer models fine-tuned on our dataset can reliably predict engagement categories in patient-nurse conversations. Lastly, we use LIME (Ribeiro et al., 2016) to analyze the underlying challenges of the tasks that state-of-the-art transformer models encounter. The de-identified data is available for research purposes upon request.
comment: ACL Findings 2023
☆ Correcting Semantic Parses with Natural Language through Dynamic Schema Encoding ACL 2023
In addressing the task of converting natural language to SQL queries, there are several semantic and syntactic challenges. It becomes increasingly important to understand and remedy the points of failure as the performance of semantic parsing systems improve. We explore semantic parse correction with natural language feedback, proposing a new solution built on the success of autoregressive decoders in text-to-SQL tasks. By separating the semantic and syntactic difficulties of the task, we show that the accuracy of text-to-SQL parsers can be boosted by up to 26% with only one turn of correction with natural language. Additionally, we show that a T5-base model is capable of correcting the errors of a T5-large model in a zero-shot, cross-parser setting.
comment: ACL 2023 Workshop on NLP for Conversational AI
☆ ViLaS: Integrating Vision and Language into Automatic Speech Recognition
Employing additional multimodal information to improve automatic speech recognition (ASR) performance has been proven effective in previous works. However, many of these works focus only on the utilization of visual cues from human lip motion. In fact, context-dependent visual and linguistic cues can also be used to improve ASR performance in many scenarios. In this paper, we first propose a multimodal ASR model (ViLaS) that can simultaneously or separately integrate visual and linguistic cues to help recognize the input speech, and introduce a training strategy that can improve performance in modal-incomplete test scenarios. Then, we create a multimodal ASR dataset (VSDial) with visual and linguistic cues to explore the effects of integrating vision and language. Finally, we report empirical results on the public Flickr8K and self-constructed VSDial datasets, investigate cross-modal fusion schemes, and analyze fine-grained cross-modal alignment on VSDial.
comment: Speech Recognition with visual-context inputs, Multimodal Speech Recognition, 5 pages
☆ Metropolis-Hastings algorithm in joint-attention naming game: Experimental semiotics study
In this study, we explore the emergence of symbols during interactions between individuals through an experimental semiotic study. Previous studies investigate how humans organize symbol systems through communication using artificially designed subjective experiments. In this study, we have focused on a joint attention-naming game (JA-NG) in which participants independently categorize objects and assign names while assuming their joint attention. In the theory of the Metropolis-Hastings naming game (MHNG), listeners accept provided names according to the acceptance probability computed using the Metropolis-Hastings (MH) algorithm. The theory of MHNG suggests that symbols emerge as an approximate decentralized Bayesian inference of signs, which is represented as a shared prior variable if the conditions of MHNG are satisfied. This study examines whether human participants exhibit behavior consistent with MHNG theory when playing JA-NG. By comparing human acceptance decisions of a partner's naming with acceptance probabilities computed in the MHNG, we tested whether human behavior is consistent with the MHNG theory. The main contributions of this study are twofold. First, we reject the null hypothesis that humans make acceptance judgments with a constant probability, regardless of the acceptance probability calculated by the MH algorithm. This result suggests that people followed the acceptance probability computed by the MH algorithm to some extent. Second, the MH-based model predicted human acceptance/rejection behavior more accurately than the other four models: Constant, Numerator, Subtraction, and Binary. This result indicates that symbol emergence in JA-NG can be explained using MHNG and is considered an approximate decentralized Bayesian inference.
☆ Speaking the Language of Your Listener: Audience-Aware Adaptation via Plug-and-Play Theory of Mind ACL 2023
Dialogue participants may have varying levels of knowledge about the topic under discussion. In such cases, it is essential for speakers to adapt their utterances by taking their audience into account. Yet, it is an open question how such adaptation can be modelled in computational agents. In this paper, we model a visually grounded referential game between a knowledgeable speaker and a listener with more limited visual and linguistic experience. Inspired by psycholinguistic theories, we endow our speaker with the ability to adapt its referring expressions via a simulation module that monitors the effectiveness of planned utterances from the listener's perspective. We propose an adaptation mechanism building on plug-and-play approaches to controlled language generation, where utterance generation is steered on the fly by the simulator without finetuning the speaker's underlying language model. Our results and analyses show that our approach is effective: the speaker's utterances become closer to the listener's domain of expertise, which leads to higher communicative success.
comment: To appear in Findings of ACL 2023
☆ A Global Context Mechanism for Sequence Labeling
Sequential labeling tasks necessitate the computation of sentence representations for each word within a given sentence. With the advent of advanced pretrained language models; one common approach involves incorporating a BiLSTM layer to bolster the sequence structure information at the output level. Nevertheless, it has been empirically demonstrated (P.-H. Li et al., 2020) that the potential of BiLSTM for generating sentence representations for sequence labeling tasks is constrained, primarily due to the amalgamation of fragments form past and future sentence representations to form a complete sentence representation. In this study, we discovered that strategically integrating the whole sentence representation, which existing in the first cell and last cell of BiLSTM, into sentence representation of ecah cell, could markedly enhance the F1 score and accuracy. Using BERT embedded within BiLSTM as illustration, we conducted exhaustive experiments on nine datasets for sequence labeling tasks, encompassing named entity recognition (NER), part of speech (POS) tagging and End-to-End Aspect-Based sentiment analysis (E2E-ABSA). We noted significant improvements in F1 scores and accuracy across all examined datasets .
☆ ChatGPT an ENFJ, Bard an ISTJ: Empirical Study on Personalities of Large Language Models
Large Language Models (LLMs) have made remarkable advancements in the field of artificial intelligence, significantly reshaping the human-computer interaction. We not only focus on the performance of LLMs, but also explores their features from a psychological perspective, acknowledging the importance of understanding their behavioral characteristics. Our study examines the behavioral patterns displayed by LLMs by employing trait theory, a psychological framework. We first focus on evaluating the consistency of personality types exhibited by ChatGPT. Furthermore, experiments include cross-lingual effects on seven additional languages, and the investigation of four other LLMs. Moreover, the study investigates whether ChatGPT can exhibit personality changes in response to instructions or contextual cues. The findings show that ChatGPT consistently maintains its ENFJ personality regardless of instructions or contexts. By shedding light on the personalization of LLMs, we anticipate that our study will serve as a catalyst for further research in this field.
comment: Working in progress
☆ Data Augmentation Approaches for Source Code Models: A Survey
The increasingly popular adoption of source code in many critical tasks motivates the development of data augmentation (DA) techniques to enhance training data and improve various capabilities (e.g., robustness and generalizability) of these models. Although a series of DA methods have been proposed and tailored for source code models, there lacks a comprehensive survey and examination to understand their effectiveness and implications. This paper fills this gap by conducting a comprehensive and integrative survey of data augmentation for source code, wherein we systematically compile and encapsulate existing literature to provide a comprehensive overview of the field. We start by constructing a taxonomy of DA for source code models model approaches, followed by a discussion on prominent, methodologically illustrative approaches. Next, we highlight the general strategies and techniques to optimize the DA quality. Subsequently, we underscore techniques that find utility in widely-accepted source code scenarios and downstream tasks. Finally, we outline the prevailing challenges and potential opportunities for future research. In essence, this paper endeavors to demystify the corpus of existing literature on DA for source code models, and foster further exploration in this sphere. Complementing this, we present a continually updated GitHub repository that hosts a list of update-to-date papers on DA for source code models, accessible at \url{https://github.com/terryyz/DataAug4Code}.
comment: Technical Report
☆ Neuron to Graph: Interpreting Language Model Neurons at Scale
Advances in Large Language Models (LLMs) have led to remarkable capabilities, yet their inner mechanisms remain largely unknown. To understand these models, we need to unravel the functions of individual neurons and their contribution to the network. This paper introduces a novel automated approach designed to scale interpretability techniques across a vast array of neurons within LLMs, to make them more interpretable and ultimately safe. Conventional methods require examination of examples with strong neuron activation and manual identification of patterns to decipher the concepts a neuron responds to. We propose Neuron to Graph (N2G), an innovative tool that automatically extracts a neuron's behaviour from the dataset it was trained on and translates it into an interpretable graph. N2G uses truncation and saliency methods to emphasise only the most pertinent tokens to a neuron while enriching dataset examples with diverse samples to better encompass the full spectrum of neuron behaviour. These graphs can be visualised to aid researchers' manual interpretation, and can generate token activations on text for automatic validation by comparison with the neuron's ground truth activations, which we use to show that the model is better at predicting neuron activation than two baseline methods. We also demonstrate how the generated graph representations can be flexibly used to facilitate further automation of interpretability research, by searching for neurons with particular properties, or programmatically comparing neurons to each other to identify similar neurons. Our method easily scales to build graph representations for all neurons in a 6-layer Transformer model using a single Tesla T4 GPU, allowing for wide usability. We release the code and instructions for use at https://github.com/alexjfoote/Neuron2Graph.
☆ How to Plant Trees in Language Models: Data and Architectural Effects on the Emergence of Syntactic Inductive Biases ACL 2023
Accurate syntactic representations are essential for robust generalization in natural language. Recent work has found that pre-training can teach language models to rely on hierarchical syntactic features - as opposed to incorrect linear features - when performing tasks after fine-tuning. We test what aspects of pre-training are important for endowing encoder-decoder Transformers with an inductive bias that favors hierarchical syntactic generalizations. We focus on architectural features (depth, width, and number of parameters), as well as the genre and size of the pre-training corpus, diagnosing inductive biases using two syntactic transformation tasks: question formation and passivization, both in English. We find that the number of parameters alone does not explain hierarchical generalization: model depth plays greater role than model width. We also find that pre-training on simpler language, such as child-directed speech, induces a hierarchical bias using an order-of-magnitude less data than pre-training on more typical datasets based on web text or Wikipedia; this suggests that in cognitively plausible language acquisition settings, neural language models may be more data-efficient than previously thought.
comment: Accepted to ACL 2023
☆ AQE: Argument Quadruplet Extraction via a Quad-Tagging Augmented Generative Approach ACL 2023
Argument mining involves multiple sub-tasks that automatically identify argumentative elements, such as claim detection, evidence extraction, stance classification, etc. However, each subtask alone is insufficient for a thorough understanding of the argumentative structure and reasoning process. To learn a complete view of an argument essay and capture the interdependence among argumentative components, we need to know what opinions people hold (i.e., claims), why those opinions are valid (i.e., supporting evidence), which source the evidence comes from (i.e., evidence type), and how those claims react to the debating topic (i.e., stance). In this work, we for the first time propose a challenging argument quadruplet extraction task (AQE), which can provide an all-in-one extraction of four argumentative components, i.e., claims, evidence, evidence types, and stances. To support this task, we construct a large-scale and challenging dataset. However, there is no existing method that can solve the argument quadruplet extraction. To fill this gap, we propose a novel quad-tagging augmented generative approach, which leverages a quadruplet tagging module to augment the training of the generative framework. The experimental results on our dataset demonstrate the empirical superiority of our proposed approach over several strong baselines.
comment: Findings of ACL 2023
☆ Med-UniC: Unifying Cross-Lingual Medical Vision-Language Pre-Training by Diminishing Bias
The scarcity of data presents a critical obstacle to the efficacy of medical visionlanguage pre-training (VLP). A potential solution lies in the combination of datasets from various language communities. Nevertheless, the main challenge stems from the complexity of integrating diverse syntax and semantics, language-specific medical terminology, and culture-specific implicit knowledge. Therefore, one crucial aspect to consider is the presence of community bias caused by different languages. This paper presents a novel framework named Unifying Cross-Lingual Medical Vision-Language Pre-Training (Med-UniC), designed to integrate multimodal medical data from the two most prevalent languages, English and Spanish. Specifically, we propose Cross-lingual Text Alignment Regularization (CTR) to explicitly unify cross-lingual semantic representations of medical reports originating from diverse language communities. CTR is optimized through latent language disentanglement, rendering our optimization objective to not depend on negative samples, thereby significantly mitigating the bias from determining positive-negative sample pairs within analogous medical reports. Furthermore, it ensures that the cross-lingual representation is not biased toward any specific language community. Med-UniC reaches superior performance across 5 medical image tasks and 10 datasets encompassing over 30 diseases, offering a versatile framework for unifying multi-modal medical data within diverse linguistic communities. The experimental outcomes highlight the presence of community bias in cross-lingual VLP. Reducing this bias enhances the performance not only in vision-language tasks but also in uni-modal visual tasks.
comment: Under review
☆ TPDM: Selectively Removing Positional Information for Zero-shot Translation via Token-Level Position Disentangle Module
Due to Multilingual Neural Machine Translation's (MNMT) capability of zero-shot translation, many works have been carried out to fully exploit the potential of MNMT in zero-shot translation. It is often hypothesized that positional information may hinder the MNMT from outputting a robust encoded representation for decoding. However, previous approaches treat all the positional information equally and thus are unable to selectively remove certain positional information. In sharp contrast, this paper investigates how to learn to selectively preserve useful positional information. We describe the specific mechanism of positional information influencing MNMT from the perspective of linguistics at the token level. We design a token-level position disentangle module (TPDM) framework to disentangle positional information at the token level based on the explanation. Our experiments demonstrate that our framework improves zero-shot translation by a large margin while reducing the performance loss in the supervised direction compared to previous works.
☆ How Does Pretraining Improve Discourse-Aware Translation?
Pretrained language models (PLMs) have produced substantial improvements in discourse-aware neural machine translation (NMT), for example, improved coherence in spoken language translation. However, the underlying reasons for their strong performance have not been well explained. To bridge this gap, we introduce a probing task to interpret the ability of PLMs to capture discourse relation knowledge. We validate three state-of-the-art PLMs across encoder-, decoder-, and encoder-decoder-based models. The analysis shows that (1) the ability of PLMs on discourse modelling varies from architecture and layer; (2) discourse elements in a text lead to different learning difficulties for PLMs. Besides, we investigate the effects of different PLMs on spoken language translation. Through experiments on IWSLT2017 Chinese-English dataset, we empirically reveal that NMT models initialized from different layers of PLMs exhibit the same trends with the probing task. Our findings are instructive to understand how and when discourse knowledge in PLMs should work for downstream tasks.
comment: Interspeech 2023
☆ Guiding Computational Stance Detection with Expanded Stance Triangle Framework ACL 2023
Stance detection determines whether the author of a piece of text is in favor of, against, or neutral towards a specified target, and can be used to gain valuable insights into social media. The ubiquitous indirect referral of targets makes this task challenging, as it requires computational solutions to model semantic features and infer the corresponding implications from a literal statement. Moreover, the limited amount of available training data leads to subpar performance in out-of-domain and cross-target scenarios, as data-driven approaches are prone to rely on superficial and domain-specific features. In this work, we decompose the stance detection task from a linguistic perspective, and investigate key components and inference paths in this task. The stance triangle is a generic linguistic framework previously proposed to describe the fundamental ways people express their stance. We further expand it by characterizing the relationship between explicit and implicit objects. We then use the framework to extend one single training corpus with additional annotation. Experimental results show that strategically-enriched data can significantly improve the performance on out-of-domain and cross-target evaluation.
comment: Main Conference in ACL 2023
☆ BEIR-PL: Zero Shot Information Retrieval Benchmark for the Polish Language
The BEIR dataset is a large, heterogeneous benchmark for Information Retrieval (IR) in zero-shot settings, garnering considerable attention within the research community. However, BEIR and analogous datasets are predominantly restricted to the English language. Our objective is to establish extensive large-scale resources for IR in the Polish language, thereby advancing the research in this NLP area. In this work, inspired by mMARCO and Mr.~TyDi datasets, we translated all accessible open IR datasets into Polish, and we introduced the BEIR-PL benchmark -- a new benchmark which comprises 13 datasets, facilitating further development, training and evaluation of modern Polish language models for IR tasks. We executed an evaluation and comparison of numerous IR models on the newly introduced BEIR-PL benchmark. Furthermore, we publish pre-trained open IR models for Polish language,d marking a pioneering development in this field. Additionally, the evaluation revealed that BM25 achieved significantly lower scores for Polish than for English, which can be attributed to high inflection and intricate morphological structure of the Polish language. Finally, we trained various re-ranking models to enhance the BM25 retrieval, and we compared their performance to identify their unique characteristic features. To ensure accurate model comparisons, it is necessary to scrutinise individual results rather than to average across the entire benchmark. Thus, we thoroughly analysed the outcomes of IR models in relation to each individual data subset encompassed by the BEIR benchmark. The benchmark data is available at URL {\bf https://huggingface.co/clarin-knext}.
☆ Deliberate then Generate: Enhanced Prompting Framework for Text Generation
Large language models (LLMs) have shown remarkable success across a wide range of natural language generation tasks, where proper prompt designs make great impacts. While existing prompting methods are normally restricted to providing correct information, in this paper, we encourage the model to deliberate by proposing a novel Deliberate then Generate (DTG) prompting framework, which consists of error detection instructions and candidates that may contain errors. DTG is a simple yet effective technique that can be applied to various text generation tasks with minimal modifications. We conduct extensive experiments on 20+ datasets across 7 text generation tasks, including summarization, translation, dialogue, and more. We show that DTG consistently outperforms existing prompting methods and achieves state-of-the-art performance on multiple text generation tasks. We also provide in-depth analyses to reveal the underlying mechanisms of DTG, which may inspire future research on prompting for LLMs.
☆ LMCap: Few-shot Multilingual Image Captioning by Retrieval Augmented Language Model Prompting ACL 2023
Multilingual image captioning has recently been tackled by training with large-scale machine translated data, which is an expensive, noisy, and time-consuming process. Without requiring any multilingual caption data, we propose LMCap, an image-blind few-shot multilingual captioning model that works by prompting a language model with retrieved captions. Specifically, instead of following the standard encoder-decoder paradigm, given an image, LMCap first retrieves the captions of similar images using a multilingual CLIP encoder. These captions are then combined into a prompt for an XGLM decoder, in order to generate captions in the desired language. In other words, the generation model does not directly process the image, instead processing retrieved captions. Experiments on the XM3600 dataset of geographically diverse images show that our model is competitive with fully-supervised multilingual captioning models, without requiring any supervised training on any captioning data.
comment: To appear in the Findings of ACL 2023
☆ IDAS: Intent Discovery with Abstractive Summarization ACL
Intent discovery is the task of inferring latent intents from a set of unlabeled utterances, and is a useful step towards the efficient creation of new conversational agents. We show that recent competitive methods in intent discovery can be outperformed by clustering utterances based on abstractive summaries, i.e., "labels", that retain the core elements while removing non-essential information. We contribute the IDAS approach, which collects a set of descriptive utterance labels by prompting a Large Language Model, starting from a well-chosen seed set of prototypical utterances, to bootstrap an In-Context Learning procedure to generate labels for non-prototypical utterances. The utterances and their resulting noisy labels are then encoded by a frozen pre-trained encoder, and subsequently clustered to recover the latent intents. For the unsupervised task (without any intent labels) IDAS outperforms the state-of-the-art by up to +7.42% in standard cluster metrics for the Banking, StackOverflow, and Transport datasets. For the semi-supervised task (with labels for a subset of intents) IDAS surpasses 2 recent methods on the CLINC benchmark without even using labeled data.
comment: The 5th Workshop on NLP for Conversational AI (NLP4ConvAI@ACL)
☆ Attention-Based Methods For Audio Question Answering
Audio question answering (AQA) is the task of producing natural language answers when a system is provided with audio and natural language questions. In this paper, we propose neural network architectures based on self-attention and cross-attention for the AQA task. The self-attention layers extract powerful audio and textual representations. The cross-attention maps audio features that are relevant to the textual features to produce answers. All our models are trained on the recently proposed Clotho-AQA dataset for both binary yes/no questions and single-word answer questions. Our results clearly show improvement over the reference method reported in the original paper. On the yes/no binary classification task, our proposed model achieves an accuracy of 68.3% compared to 62.7% in the reference model. For the single-word answers multiclass classifier, our model produces a top-1 and top-5 accuracy of 57.9% and 99.8% compared to 54.2% and 93.7% in the reference model respectively. We further discuss some of the challenges in the Clotho-AQA dataset such as the presence of the same answer word in multiple tenses, singular and plural forms, and the presence of specific and generic answers to the same question. We address these issues and present a revised version of the dataset.
☆ Recursive Metropolis-Hastings Naming Game: Symbol Emergence in a Multi-agent System based on Probabilistic Generative Models
In the studies on symbol emergence and emergent communication in a population of agents, a computational model was employed in which agents participate in various language games. Among these, the Metropolis-Hastings naming game (MHNG) possesses a notable mathematical property: symbol emergence through MHNG is proven to be a decentralized Bayesian inference of representations shared by the agents. However, the previously proposed MHNG is limited to a two-agent scenario. This paper extends MHNG to an N-agent scenario. The main contributions of this paper are twofold: (1) we propose the recursive Metropolis-Hastings naming game (RMHNG) as an N-agent version of MHNG and demonstrate that RMHNG is an approximate Bayesian inference method for the posterior distribution over a latent variable shared by agents, similar to MHNG; and (2) we empirically evaluate the performance of RMHNG on synthetic and real image data, enabling multiple agents to develop and share a symbol system. Furthermore, we introduce two types of approximations -- one-sample and limited-length -- to reduce computational complexity while maintaining the ability to explain communication in a population of agents. The experimental findings showcased the efficacy of RMHNG as a decentralized Bayesian inference for approximating the posterior distribution concerning latent variables, which are jointly shared among agents, akin to MHNG. Moreover, the utilization of RMHNG elucidated the agents' capacity to exchange symbols. Furthermore, the study discovered that even the computationally simplified version of RMHNG could enable symbols to emerge among the agents.
☆ Simple yet Effective Code-Switching Language Identification with Multitask Pre-Training and Transfer Learning
Code-switching, also called code-mixing, is the linguistics phenomenon where in casual settings, multilingual speakers mix words from different languages in one utterance. Due to its spontaneous nature, code-switching is extremely low-resource, which makes it a challenging problem for language and speech processing tasks. In such contexts, Code-Switching Language Identification (CSLID) becomes a difficult but necessary task if we want to maximally leverage existing monolingual tools for other tasks. In this work, we propose two novel approaches toward improving language identification accuracy on an English-Mandarin child-directed speech dataset. Our methods include a stacked Residual CNN+GRU model and a multitask pre-training approach to use Automatic Speech Recognition (ASR) as an auxiliary task for CSLID. Due to the low-resource nature of code-switching, we also employ careful silver data creation using monolingual corpora in both languages and up-sampling as data augmentation. We focus on English-Mandarin code-switched data, but our method works on any language pair. Our best model achieves a balanced accuracy of 0.781 on a real English-Mandarin code-switching child-directed speech corpus and outperforms the previous baseline by 55.3%.
comment: 8 pages, 3 figures, 7 tables
☆ Automatic Discrimination of Human and Neural Machine Translation in Multilingual Scenarios
We tackle the task of automatically discriminating between human and machine translations. As opposed to most previous work, we perform experiments in a multilingual setting, considering multiple languages and multilingual pretrained language models. We show that a classifier trained on parallel data with a single source language (in our case German-English) can still perform well on English translations that come from different source languages, even when the machine translations were produced by other systems than the one it was trained on. Additionally, we demonstrate that incorporating the source text in the input of a multilingual classifier improves (i) its accuracy and (ii) its robustness on cross-system evaluation, compared to a monolingual classifier. Furthermore, we find that using training data from multiple source languages (German, Russian, and Chinese) tends to improve the accuracy of both monolingual and multilingual classifiers. Finally, we show that bilingual classifiers and classifiers trained on multiple source languages benefit from being trained on longer text sequences, rather than on sentences.
comment: Accepted at EAMT2023
☆ Sentence Simplification Using Paraphrase Corpus for Initialization
Neural sentence simplification method based on sequence-to-sequence framework has become the mainstream method for sentence simplification (SS) task. Unfortunately, these methods are currently limited by the scarcity of parallel SS corpus. In this paper, we focus on how to reduce the dependence on parallel corpus by leveraging a careful initialization for neural SS methods from paraphrase corpus. Our work is motivated by the following two findings: (1) Paraphrase corpus includes a large proportion of sentence pairs belonging to SS corpus. (2) We can construct large-scale pseudo parallel SS data by keeping these sentence pairs with a higher complexity difference. Therefore, we propose two strategies to initialize neural SS methods using paraphrase corpus. We train three different neural SS methods with our initialization, which can obtain substantial improvements on the available WikiLarge data compared with themselves without initialization.
comment: arXiv admin note: substantial text overlap with arXiv:2109.00165
☆ Text-to-Speech Pipeline for Swiss German -- A comparison
In this work, we studied the synthesis of Swiss German speech using different Text-to-Speech (TTS) models. We evaluated the TTS models on three corpora, and we found, that VITS models performed best, hence, using them for further testing. We also introduce a new method to evaluate TTS models by letting the discriminator of a trained vocoder GAN model predict whether a given waveform is human or synthesized. In summary, our best model delivers speech synthesis for different Swiss German dialects with previously unachieved quality.
☆ UKP-SQuARE: An Interactive Tool for Teaching Question Answering ACL2023
The exponential growth of question answering (QA) has made it an indispensable topic in any Natural Language Processing (NLP) course. Additionally, the breadth of QA derived from this exponential growth makes it an ideal scenario for teaching related NLP topics such as information retrieval, explainability, and adversarial attacks among others. In this paper, we introduce UKP-SQuARE as a platform for QA education. This platform provides an interactive environment where students can run, compare, and analyze various QA models from different perspectives, such as general behavior, explainability, and robustness. Therefore, students can get a first-hand experience in different QA techniques during the class. Thanks to this, we propose a learner-centered approach for QA education in which students proactively learn theoretical concepts and acquire problem-solving skills through interactive exploration, experimentation, and practical assignments, rather than solely relying on traditional lectures. To evaluate the effectiveness of UKP-SQuARE in teaching scenarios, we adopted it in a postgraduate NLP course and surveyed the students after the course. Their positive feedback shows the platform's effectiveness in their course and invites a wider adoption.
comment: Accepted by BEA workshop, ACL2023
☆ Analyzing Text Representations by Measuring Task Alignment
Textual representations based on pre-trained language models are key, especially in few-shot learning scenarios. What makes a representation good for text classification? Is it due to the geometric properties of the space or because it is well aligned with the task? We hypothesize the second claim. To test it, we develop a task alignment score based on hierarchical clustering that measures alignment at different levels of granularity. Our experiments on text classification validate our hypothesis by showing that task alignment can explain the classification performance of a given representation.
comment: arXiv admin note: text overlap with arXiv:2210.05721
☆ Knowledge Base Question Answering for Space Debris Queries ACL 2023
Space agencies execute complex satellite operations that need to be supported by the technical knowledge contained in their extensive information systems. Knowledge bases (KB) are an effective way of storing and accessing such information at scale. In this work we present a system, developed for the European Space Agency (ESA), that can answer complex natural language queries, to support engineers in accessing the information contained in a KB that models the orbital space debris environment. Our system is based on a pipeline which first generates a sequence of basic database operations, called a %program sketch, from a natural language question, then specializes the sketch into a concrete query program with mentions of entities, attributes and relations, and finally executes the program against the database. This pipeline decomposition approach enables us to train the system by leveraging out-of-domain data and semi-synthetic data generated by GPT-3, thus reducing overfitting and shortcut learning even with limited amount of in-domain training data. Our code can be found at \url{https://github.com/PaulDrm/DISCOSQA}.
comment: 7 pages, ACL 2023 industry track
☆ Red Teaming Language Model Detectors with Language Models
The prevalence and high capacity of large language models (LLMs) present significant safety and ethical risks when malicious users exploit them for automated content generation. To prevent the potentially deceptive usage of LLMs, recent works have proposed several algorithms to detect machine-generated text. In this paper, we systematically test the reliability of the existing detectors, by designing two types of attack strategies to fool the detectors: 1) replacing words with their synonyms based on the context; 2) altering the writing style of generated text. These strategies are implemented by instructing LLMs to generate synonymous word substitutions or writing directives that modify the style without human involvement, and the LLMs leveraged in the attack can also be protected by detectors. Our research reveals that our attacks effectively compromise the performance of all tested detectors, thereby underscoring the urgent need for the development of more robust machine-generated text detection systems.
comment: Work in progress. Zhouxing Shi, Yihan Wang and Fan Yin are ordered alphabetically
☆ XPhoneBERT: A Pre-trained Multilingual Model for Phoneme Representations for Text-to-Speech INTERSPEECH 2023
We present XPhoneBERT, the first multilingual model pre-trained to learn phoneme representations for the downstream text-to-speech (TTS) task. Our XPhoneBERT has the same model architecture as BERT-base, trained using the RoBERTa pre-training approach on 330M phoneme-level sentences from nearly 100 languages and locales. Experimental results show that employing XPhoneBERT as an input phoneme encoder significantly boosts the performance of a strong neural TTS model in terms of naturalness and prosody and also helps produce fairly high-quality speech with limited training data. We publicly release our pre-trained XPhoneBERT with the hope that it would facilitate future research and downstream TTS applications for multiple languages. Our XPhoneBERT model is available at https://github.com/VinAIResearch/XPhoneBERT
comment: In Proceedings of INTERSPEECH 2023 (to appear)
☆ Building Extractive Question Answering System to Support Human-AI Health Coaching Model for Sleep Domain
Non-communicable diseases (NCDs) are a leading cause of global deaths, necessitating a focus on primary prevention and lifestyle behavior change. Health coaching, coupled with Question Answering (QA) systems, has the potential to transform preventive healthcare. This paper presents a human-Artificial Intelligence (AI) health coaching model incorporating a domain-specific extractive QA system. A sleep-focused dataset, SleepQA, was manually assembled and used to fine-tune domain-specific BERT models. The QA system was evaluated using automatic and human methods. A data-centric framework enhanced the system's performance by improving passage retrieval and question reformulation. Although the system did not outperform the baseline in automatic evaluation, it excelled in the human evaluation of real-world questions. Integration into a Human-AI health coaching model was tested in a pilot Randomized Controlled Trial (RCT).
comment: 2 pages, 1 figure
☆ Assessing Word Importance Using Models Trained for Semantic Tasks ACL 2023
Many NLP tasks require to automatically identify the most significant words in a text. In this work, we derive word significance from models trained to solve semantic task: Natural Language Inference and Paraphrase Identification. Using an attribution method aimed to explain the predictions of these models, we derive importance scores for each input token. We evaluate their relevance using a so-called cross-task evaluation: Analyzing the performance of one model on an input masked according to the other model's weight, we show that our method is robust with respect to the choice of the initial task. Additionally, we investigate the scores from the syntax point of view and observe interesting patterns, e.g. words closer to the root of a syntactic tree receive higher importance scores. Altogether, these observations suggest that our method can be used to identify important words in sentences without any explicit word importance labeling in training.
comment: Published in the Findings of ACL 2023
☆ Unveiling Cross Modality Bias in Visual Question Answering: A Causal View with Possible Worlds VQA
To increase the generalization capability of VQA systems, many recent studies have tried to de-bias spurious language or vision associations that shortcut the question or image to the answer. Despite these efforts, the literature fails to address the confounding effect of vision and language simultaneously. As a result, when they reduce bias learned from one modality, they usually increase bias from the other. In this paper, we first model a confounding effect that causes language and vision bias simultaneously, then propose a counterfactual inference to remove the influence of this effect. The model trained in this strategy can concurrently and efficiently reduce vision and language bias. To the best of our knowledge, this is the first work to reduce biases resulting from confounding effects of vision and language in VQA, leveraging causal explain-away relations. We accompany our method with an explain-away strategy, pushing the accuracy of the questions with numerical answers results compared to existing methods that have been an open problem. The proposed method outperforms the state-of-the-art methods in VQA-CP v2 datasets.
comment: 22 pages
☆ Adverbs, Surprisingly
This paper begins with the premise that adverbs are neglected in computational linguistics. This view derives from two analyses: a literature review and a novel adverb dataset to probe a state-of-the-art language model, thereby uncovering systematic gaps in accounts for adverb meaning. We suggest that using Frame Semantics for characterizing word meaning, as in FrameNet, provides a promising approach to adverb analysis, given its ability to describe ambiguity, semantic roles, and null instantiation.
☆ Adversarial Clean Label Backdoor Attacks and Defenses on Text Classification Systems RepL4NLP 2023
Clean-label (CL) attack is a form of data poisoning attack where an adversary modifies only the textual input of the training data, without requiring access to the labeling function. CL attacks are relatively unexplored in NLP, as compared to label flipping (LF) attacks, where the latter additionally requires access to the labeling function as well. While CL attacks are more resilient to data sanitization and manual relabeling methods than LF attacks, they often demand as high as ten times the poisoning budget than LF attacks. In this work, we first introduce an Adversarial Clean Label attack which can adversarially perturb in-class training examples for poisoning the training set. We then show that an adversary can significantly bring down the data requirements for a CL attack, using the aforementioned approach, to as low as 20% of the data otherwise required. We then systematically benchmark and analyze a number of defense methods, for both LF and CL attacks, some previously employed solely for LF attacks in the textual domain and others adapted from computer vision. We find that text-specific defenses greatly vary in their effectiveness depending on their properties.
comment: RepL4NLP 2023 at ACL 2023
☆ What does the Failure to Reason with "Respectively" in Zero/Few-Shot Settings Tell Us about Language Models? ACL 2023
Humans can effortlessly understand the coordinate structure of sentences such as "Niels Bohr and Kurt Cobain were born in Copenhagen and Seattle, respectively". In the context of natural language inference (NLI), we examine how language models (LMs) reason with respective readings (Gawron and Kehler, 2004) from two perspectives: syntactic-semantic and commonsense-world knowledge. We propose a controlled synthetic dataset WikiResNLI and a naturally occurring dataset NatResNLI to encompass various explicit and implicit realizations of "respectively". We show that fine-tuned NLI models struggle with understanding such readings without explicit supervision. While few-shot learning is easy in the presence of explicit cues, longer training is required when the reading is evoked implicitly, leaving models to rely on common sense inferences. Furthermore, our fine-grained analysis indicates models fail to generalize across different constructions. To conclude, we demonstrate that LMs still lag behind humans in generalizing to the long tail of linguistic constructions.
comment: To appear at ACL 2023
☆ SLABERT Talk Pretty One Day: Modeling Second Language Acquisition with BERT ACL 2023
Second language acquisition (SLA) research has extensively studied cross-linguistic transfer, the influence of linguistic structure of a speaker's native language [L1] on the successful acquisition of a foreign language [L2]. Effects of such transfer can be positive (facilitating acquisition) or negative (impeding acquisition). We find that NLP literature has not given enough attention to the phenomenon of negative transfer. To understand patterns of both positive and negative transfer between L1 and L2, we model sequential second language acquisition in LMs. Further, we build a Mutlilingual Age Ordered CHILDES (MAO-CHILDES) -- a dataset consisting of 5 typologically diverse languages, i.e., German, French, Polish, Indonesian, and Japanese -- to understand the degree to which native Child-Directed Speech (CDS) [L1] can help or conflict with English language acquisition [L2]. To examine the impact of native CDS, we use the TILT-based cross lingual transfer learning approach established by Papadimitriou and Jurafsky (2020) and find that, as in human SLA, language family distance predicts more negative transfer. Additionally, we find that conversational speech data shows greater facilitation for language acquisition than scripted speech data. Our findings call for further research using our novel Transformer-based SLA models and we would like to encourage it by releasing our code, data, and models.
comment: Accepted at ACL 2023
☆ LAIT: Efficient Multi-Segment Encoding in Transformers with Layer-Adjustable Interaction ACL 2023
Transformer encoders contextualize token representations by attending to all other tokens at each layer, leading to quadratic increase in compute effort with the input length. In practice, however, the input text of many NLP tasks can be seen as a sequence of related segments (e.g., the sequence of sentences within a passage, or the hypothesis and premise in NLI). While attending across these segments is highly beneficial for many tasks, we hypothesize that this interaction can be delayed until later encoding stages. To this end, we introduce Layer-Adjustable Interactions in Transformers (LAIT). Within LAIT, segmented inputs are first encoded independently, and then jointly. This partial two-tower architecture bridges the gap between a Dual Encoder's ability to pre-compute representations for segments and a fully self-attentive Transformer's capacity to model cross-segment attention. The LAIT framework effectively leverages existing pretrained Transformers and converts them into the hybrid of the two aforementioned architectures, allowing for easy and intuitive control over the performance-efficiency tradeoff. Experimenting on a wide range of NLP tasks, we find LAIT able to reduce 30-50% of the attention FLOPs on many tasks, while preserving high accuracy; in some practical settings, LAIT could reduce actual latency by orders of magnitude.
comment: ACL 2023
☆ The Tag-Team Approach: Leveraging CLS and Language Tagging for Enhancing Multilingual ASR INTERSPEECH2023
Building a multilingual Automated Speech Recognition (ASR) system in a linguistically diverse country like India can be a challenging task due to the differences in scripts and the limited availability of speech data. This problem can be solved by exploiting the fact that many of these languages are phonetically similar. These languages can be converted into a Common Label Set (CLS) by mapping similar sounds to common labels. In this paper, new approaches are explored and compared to improve the performance of CLS based multilingual ASR model. Specific language information is infused in the ASR model by giving Language ID or using CLS to Native script converter on top of the CLS Multilingual model. These methods give a significant improvement in Word Error Rate (WER) compared to the CLS baseline. These methods are further tried on out-of-distribution data to check their robustness.
comment: 5 pages,5 figures, submitted to INTERSPEECH2023
☆ DC CoMix TTS: An End-to-End Expressive TTS with Discrete Code Collaborated with Mixer
Despite the huge successes made in neutral TTS, content-leakage remains a challenge. In this paper, we propose a new input representation and simple architecture to achieve improved prosody modeling. Inspired by the recent success in the use of discrete code in TTS, we introduce discrete code to the input of the reference encoder. Specifically, we leverage the vector quantizer from the audio compression model to exploit the diverse acoustic information it has already been trained on. In addition, we apply the modified MLP-Mixer to the reference encoder, making the architecture lighter. As a result, we train the prosody transfer TTS in an end-to-end manner. We prove the effectiveness of our method through both subjective and objective evaluations. We demonstrate that the reference encoder learns better speaker-independent prosody when discrete code is utilized as input in the experiments. In addition, we obtain comparable results even when fewer parameters are inputted.
comment: Accepted at Interspeech 2023
☆ Zero-Shot Automatic Pronunciation Assessment
Automatic Pronunciation Assessment (APA) is vital for computer-assisted language learning. Prior methods rely on annotated speech-text data to train Automatic Speech Recognition (ASR) models or speech-score data to train regression models. In this work, we propose a novel zero-shot APA method based on the pre-trained acoustic model, HuBERT. Our method involves encoding speech input and corrupting them via a masking module. We then employ the Transformer encoder and apply k-means clustering to obtain token sequences. Finally, a scoring module is designed to measure the number of wrongly recovered tokens. Experimental results on speechocean762 demonstrate that the proposed method achieves comparable performance to supervised regression baselines and outperforms non-regression baselines in terms of Pearson Correlation Coefficient (PCC). Additionally, we analyze how masking strategies affect the performance of APA.
comment: Accepted to Interspeech 2023
☆ Large Language Models Are Not Abstract Reasoners
Large Language Models have shown tremendous performance on a large variety of natural language processing tasks, ranging from text comprehension to common sense reasoning. However, the mechanisms responsible for this success remain unknown, and it is unclear whether LLMs can achieve human-like cognitive capabilities or whether these models are still fundamentally limited. Abstract reasoning is a fundamental task for cognition, consisting of finding and applying a general pattern from few data. Evaluating deep neural architectures on this task could give insight into their potential limitations regarding reasoning and their broad generalisation abilities, yet this is currently an under-explored area. In this paper, we perform extensive evaluations of state-of-the-art LLMs on abstract reasoning tasks, showing that they achieve very limited performance in contrast with other natural language tasks, and we investigate the reasons for this difference. We apply techniques that have been shown to improve performance on other NLP tasks and show that in most cases their impact on abstract reasoning performance is limited. In the course of this work, we have generated a new benchmark for evaluating language models on abstract reasoning tasks.
comment: 33 pages, 13 pages for the main paper and 20 pages for the supplement, 26 figures
☆ Accurate and Structured Pruning for Efficient Automatic Speech Recognition INTERSPEECH 2023
Automatic Speech Recognition (ASR) has seen remarkable advancements with deep neural networks, such as Transformer and Conformer. However, these models typically have large model sizes and high inference costs, posing a challenge to deploy on resource-limited devices. In this paper, we propose a novel compression strategy that leverages structured pruning and knowledge distillation to reduce the model size and inference cost of the Conformer model while preserving high recognition performance. Our approach utilizes a set of binary masks to indicate whether to retain or prune each Conformer module, and employs L0 regularization to learn the optimal mask values. To further enhance pruning performance, we use a layerwise distillation strategy to transfer knowledge from unpruned to pruned models. Our method outperforms all pruning baselines on the widely used LibriSpeech benchmark, achieving a 50% reduction in model size and a 28% reduction in inference cost with minimal performance loss.
comment: Accepted at INTERSPEECH 2023
☆ Fine-grained Text Style Transfer with Diffusion-Based Language Models ACL 2023
Diffusion probabilistic models have shown great success in generating high-quality images controllably, and researchers have tried to utilize this controllability into text generation domain. Previous works on diffusion-based language models have shown that they can be trained without external knowledge (such as pre-trained weights) and still achieve stable performance and controllability. In this paper, we trained a diffusion-based model on StylePTB dataset, the standard benchmark for fine-grained text style transfers. The tasks in StylePTB requires much more refined control over the output text compared to tasks evaluated in previous works, and our model was able to achieve state-of-the-art performance on StylePTB on both individual and compositional transfers. Moreover, our model, trained on limited data from StylePTB without external knowledge, outperforms previous works that utilized pretrained weights, embeddings, and external grammar parsers, and this may indicate that diffusion-based language models have great potential under low-resource settings.
comment: Accepted at Repl4NLP workshop at ACL 2023
☆ Exploring Lottery Prompts for Pre-trained Language Models ACL 2023
Consistently scaling pre-trained language models (PLMs) imposes substantial burdens on model adaptation, necessitating more efficient alternatives to conventional fine-tuning. Given the advantage of prompting in the zero-shot setting and the observed performance fluctuation among different prompts, we explore the instance-level prompt and their generalizability. By searching through the prompt space, we first validate the assumption that for every instance, there is almost always a lottery prompt that induces the correct prediction from the PLM, and such prompt can be obtained at a low cost thanks to the inherent ability of PLMs. Meanwhile, we find that some strong lottery prompts have high performance over the whole training set, and they are equipped with distinguishable linguistic features. Lastly, we attempt to generalize the searched strong lottery prompts to unseen data with prompt ensembling method without any parameter tuning. Experiments are conducted on various types of NLP classification tasks and demonstrate that the proposed method can achieve comparable results with other gradient-free and optimization-free baselines.
comment: Accepted to ACL 2023
☆ Towards Flow Graph Prediction of Open-Domain Procedural Texts RepL4NLP 2023
Machine comprehension of procedural texts is essential for reasoning about the steps and automating the procedures. However, this requires identifying entities within a text and resolving the relationships between the entities. Previous work focused on the cooking domain and proposed a framework to convert a recipe text into a flow graph (FG) representation. In this work, we propose a framework based on the recipe FG for flow graph prediction of open-domain procedural texts. To investigate flow graph prediction performance in non-cooking domains, we introduce the wikiHow-FG corpus from articles on wikiHow, a website of how-to instruction articles. In experiments, we consider using the existing recipe corpus and performing domain adaptation from the cooking to the target domain. Experimental results show that the domain adaptation models achieve higher performance than those trained only on the cooking or target domain data.
comment: RepL4NLP 2023
☆ Ethical Considerations for Machine Translation of Indigenous Languages: Giving a Voice to the Speakers ACL2023
In recent years machine translation has become very successful for high-resource language pairs. This has also sparked new interest in research on the automatic translation of low-resource languages, including Indigenous languages. However, the latter are deeply related to the ethnic and cultural groups that speak (or used to speak) them. The data collection, modeling and deploying machine translation systems thus result in new ethical questions that must be addressed. Motivated by this, we first survey the existing literature on ethical considerations for the documentation, translation, and general natural language processing for Indigenous languages. Afterward, we conduct and analyze an interview study to shed light on the positions of community leaders, teachers, and language activists regarding ethical concerns for the automatic translation of their languages. Our results show that the inclusion, at different degrees, of native speakers and community members is vital to performing better and more ethical research on Indigenous languages.
comment: Accepted to ACL2023 Main Conference
☆ PlaSma: Making Small Language Models Better Procedural Knowledge Models for (Counterfactual) Planning
Procedural planning, which entails decomposing a high-level goal into a sequence of temporally ordered steps, is an important yet intricate task for machines. It involves integrating common-sense knowledge to reason about complex contextualized situations that are often counterfactual, e.g. "scheduling a doctor's appointment without a phone". While current approaches show encouraging results using large language models (LLMs), they are hindered by drawbacks such as costly API calls and reproducibility issues. In this paper, we advocate planning using smaller language models. We present PlaSma, a novel two-pronged approach to endow small language models with procedural knowledge and (counterfactual) planning capabilities. More concretely, we develop symbolic procedural knowledge distillation to enhance the implicit knowledge in small language models and an inference-time algorithm to facilitate more structured and accurate reasoning. In addition, we introduce a novel task, Counterfactual Planning, that requires a revision of a plan to cope with a counterfactual situation. In both the original and counterfactual setting, we show that orders-of-magnitude smaller models (770M-11B parameters) can compete and often surpass their larger teacher models' capabilities.
comment: 27 pages
☆ The Impact of Positional Encoding on Length Generalization in Transformers
Length generalization, the ability to generalize from small training context sizes to larger ones, is a critical challenge in the development of Transformer-based language models. Positional encoding (PE) has been identified as a major factor influencing length generalization, but the exact impact of different PE schemes on extrapolation in downstream tasks remains unclear. In this paper, we conduct a systematic empirical study comparing the length generalization performance of decoder-only Transformers with five different position encoding approaches including Absolute Position Embedding (APE), T5's Relative PE, ALiBi, and Rotary, in addition to Transformers without positional encoding (NoPE). Our evaluation encompasses a battery of reasoning and mathematical tasks. Our findings reveal that the most commonly used positional encoding methods, such as ALiBi, Rotary, and APE, are not well suited for length generalization in downstream tasks. More importantly, NoPE outperforms other explicit positional encoding methods while requiring no additional computation. We theoretically demonstrate that NoPE can represent both absolute and relative PEs, but when trained with SGD, it mostly resembles T5's relative PE attention patterns. Finally, we find that scratchpad is not always helpful to solve length generalization and its format highly impacts the model's performance. Overall, our work suggests that explicit position embeddings are not essential for decoder-only Transformers to generalize well to longer sequences.
☆ From Pixels to UI Actions: Learning to Follow Instructions via Graphical User Interfaces
Much of the previous work towards digital agents for graphical user interfaces (GUIs) has relied on text-based representations (derived from HTML or other structured data sources), which are not always readily available. These input representations have been often coupled with custom, task-specific action spaces. This paper focuses on creating agents that interact with the digital world using the same conceptual interface that humans commonly use -- via pixel-based screenshots and a generic action space corresponding to keyboard and mouse actions. Building upon recent progress in pixel-based pretraining, we show, for the first time, that it is possible for such agents to outperform human crowdworkers on the MiniWob++ benchmark of GUI-based instruction following tasks.
☆ Using Visual Cropping to Enhance Fine-Detail Question Answering of BLIP-Family Models
Visual Question Answering is a challenging task, as it requires seamless interaction between perceptual, linguistic, and background knowledge systems. While the recent progress of visual and natural language models like BLIP has led to improved performance on this task, we lack understanding of the ability of such models to perform on different kinds of questions and reasoning types. As our initial analysis of BLIP-family models revealed difficulty with answering fine-detail questions, we investigate the following question: Can visual cropping be employed to improve the performance of state-of-the-art visual question answering models on fine-detail questions? Given the recent success of the BLIP-family models, we study a zero-shot and a fine-tuned BLIP model. We define three controlled subsets of the popular VQA-v2 benchmark to measure whether cropping can help model performance. Besides human cropping, we devise two automatic cropping strategies based on multi-modal embedding by CLIP and BLIP visual QA model gradients. Our experiments demonstrate that the performance of BLIP model variants can be significantly improved through human cropping, and automatic cropping methods can produce comparable benefits. A deeper dive into our findings indicates that the performance enhancement is more pronounced in zero-shot models than in fine-tuned models and more salient with smaller bounding boxes than larger ones. We perform case studies to connect quantitative differences with qualitative observations across question types and datasets. Finally, we see that the cropping enhancement is robust, as we gain an improvement of 4.59% (absolute) in the general VQA-random task by simply inputting a concatenation of the original and gradient-based cropped images. We make our code available to facilitate further innovation on visual cropping methods for question answering.
comment: 16 pages, 5 figures, 7 tables
☆ Diffusion Brush: A Latent Diffusion Model-based Editing Tool for AI-generated Images
Text-to-image generative models have made remarkable advancements in generating high-quality images. However, generated images often contain undesirable artifacts or other errors due to model limitations. Existing techniques to fine-tune generated images are time-consuming (manual editing), produce poorly-integrated results (inpainting), or result in unexpected changes across the entire image (variation selection and prompt fine-tuning). In this work, we present Diffusion Brush, a Latent Diffusion Model-based (LDM) tool to efficiently fine-tune desired regions within an AI-synthesized image. Our method introduces new random noise patterns at targeted regions during the reverse diffusion process, enabling the model to efficiently make changes to the specified regions while preserving the original context for the rest of the image. We evaluate our method's usability and effectiveness through a user study with artists, comparing our technique against other state-of-the-art image inpainting techniques and editing software for fine-tuning AI-generated imagery.
☆ FEED PETs: Further Experimentation and Expansion on the Disambiguation of Potentially Euphemistic Terms
Transformers have been shown to work well for the task of English euphemism disambiguation, in which a potentially euphemistic term (PET) is classified as euphemistic or non-euphemistic in a particular context. In this study, we expand on the task in two ways. First, we annotate PETs for vagueness, a linguistic property associated with euphemisms, and find that transformers are generally better at classifying vague PETs, suggesting linguistic differences in the data that impact performance. Second, we present novel euphemism corpora in three different languages: Yoruba, Spanish, and Mandarin Chinese. We perform euphemism disambiguation experiments in each language using multilingual transformer models mBERT and XLM-RoBERTa, establishing preliminary results from which to launch future work.
☆ Strategies for improving low resource speech to text translation relying on pre-trained ASR models
This paper presents techniques and findings for improving the performance of low-resource speech to text translation (ST). We conducted experiments on both simulated and real-low resource setups, on language pairs English - Portuguese, and Tamasheq - French respectively. Using the encoder-decoder framework for ST, our results show that a multilingual automatic speech recognition system acts as a good initialization under low-resource scenarios. Furthermore, using the CTC as an additional objective for translation during training and decoding helps to reorder the internal representations and improves the final translation. Through our experiments, we try to identify various factors (initializations, objectives, and hyper-parameters) that contribute the most for improvements in low-resource setups. With only 300 hours of pre-training data, our model achieved 7.3 BLEU score on Tamasheq - French data, outperforming prior published works from IWSLT 2022 by 1.6 points.
☆ An Invariant Learning Characterization of Controlled Text Generation ACL 2023
Controlled generation refers to the problem of creating text that contains stylistic or semantic attributes of interest. Many approaches reduce this problem to training a predictor of the desired attribute. For example, researchers hoping to deploy a large language model to produce non-toxic content may use a toxicity classifier to filter generated text. In practice, the generated text to classify, which is determined by user prompts, may come from a wide range of distributions. In this paper, we show that the performance of controlled generation may be poor if the distributions of text in response to user prompts differ from the distribution the predictor was trained on. To address this problem, we cast controlled generation under distribution shift as an invariant learning problem: the most effective predictor should be invariant across multiple text environments. We then discuss a natural solution that arises from this characterization and propose heuristics for selecting natural environments. We study this characterization and the proposed method empirically using both synthetic and real data. Experiments demonstrate both the challenge of distribution shift in controlled generation and the potential of invariance methods in this setting.
comment: To appear in the 2023 Conference of the Association for Computational Linguistics (ACL 2023)
☆ Factually Consistent Summarization via Reinforcement Learning with Textual Entailment Feedback ACL 2023
Despite the seeming success of contemporary grounded text generation systems, they often tend to generate factually inconsistent text with respect to their input. This phenomenon is emphasized in tasks like summarization, in which the generated summaries should be corroborated by their source article. In this work, we leverage recent progress on textual entailment models to directly address this problem for abstractive summarization systems. We use reinforcement learning with reference-free, textual entailment rewards to optimize for factual consistency and explore the ensuing trade-offs, as improved consistency may come at the cost of less informative or more extractive summaries. Our results, according to both automatic metrics and human evaluation, show that our method considerably improves the faithfulness, salience, and conciseness of the generated summaries.
comment: ACL 2023
☆ Contrastive Hierarchical Discourse Graph for Scientific Document Summarization ACL 2023
The extended structural context has made scientific paper summarization a challenging task. This paper proposes CHANGES, a contrastive hierarchical graph neural network for extractive scientific paper summarization. CHANGES represents a scientific paper with a hierarchical discourse graph and learns effective sentence representations with dedicated designed hierarchical graph information aggregation. We also propose a graph contrastive learning module to learn global theme-aware sentence representations. Extensive experiments on the PubMed and arXiv benchmark datasets prove the effectiveness of CHANGES and the importance of capturing hierarchical structure information in modeling scientific papers.
comment: CODI at ACL 2023
☆ Automated Annotation with Generative AI Requires Validation
Generative large language models (LLMs) can be a powerful tool for augmenting text annotation procedures, but their performance varies across annotation tasks due to prompt quality, text data idiosyncrasies, and conceptual difficulty. Because these challenges will persist even as LLM technology improves, we argue that any automated annotation process using an LLM must validate the LLM's performance against labels generated by humans. To this end, we outline a workflow to harness the annotation potential of LLMs in a principled, efficient way. Using GPT-4, we validate this approach by replicating 27 annotation tasks across 11 datasets from recent social science articles in high-impact journals. We find that LLM performance for text annotation is promising but highly contingent on both the dataset and the type of annotation task, which reinforces the necessity to validate on a task-by-task basis. We make available easy-to-use software designed to implement our workflow and streamline the deployment of LLMs for automated annotation.
☆ Measuring the Robustness of Natural Language Processing Models to Domain Shifts
Large Language Models have shown promising performance on various tasks, including fine-tuning, few-shot learning, and zero-shot learning. However, their performance on domains without labeled data still lags behind those with labeled data, which we refer as the Domain Robustness (DR) challenge. Existing research on DR suffers from disparate setups, lack of evaluation task variety, and reliance on challenge sets. In this paper, we explore the DR challenge of both fine-tuned and few-shot learning models in natural domain shift settings. We introduce a DR benchmark comprising diverse NLP tasks, including sentence and token-level classification, QA, and generation, each task consists of several domains. We propose two views of the DR challenge: Source Drop (SD) and Target Drop (TD), which alternate between the source and target in-domain performance as reference points. We find that in significant proportions of domain shifts, either SD or TD is positive, but not both, emphasizing the importance of considering both measures as diagnostic tools. Our experimental results demonstrate the persistent existence of the DR challenge in both fine-tuning and few-shot learning models, though it is less pronounced in the latter. We also find that increasing the fine-tuned model size improves performance, particularly in classification.
☆ A Sequence-to-Sequence&Set Model for Text-to-Table Generation
Recently, the text-to-table generation task has attracted increasing attention due to its wide applications. In this aspect, the dominant model formalizes this task as a sequence-to-sequence generation task and serializes each table into a token sequence during training by concatenating all rows in a top-down order. However, it suffers from two serious defects: 1) the predefined order introduces a wrong bias during training, which highly penalizes shifts in the order between rows; 2) the error propagation problem becomes serious when the model outputs a long token sequence. In this paper, we first conduct a preliminary study to demonstrate the generation of most rows is order-insensitive. Furthermore, we propose a novel sequence-to-sequence&set text-to-table generation model. Specifically, in addition to a text encoder encoding the input text, our model is equipped with a table header generator to first output a table header, i.e., the first row of the table, in the manner of sequence generation. Then we use a table body generator with learnable row embeddings and column embeddings to generate a set of table body rows in parallel. Particularly, to deal with the issue that there is no correspondence between each generated table body row and target during training, we propose a target assignment strategy based on the bipartite matching between the first cells of generated table body rows and targets. Experiment results show that our model significantly surpasses the baselines, achieving state-of-the-art performance on commonly-used datasets.
Pre-Trained Language-Meaning Models for Multilingual Parsing and Generation ACL2023
Pre-trained language models (PLMs) have achieved great success in NLP and have recently been used for tasks in computational semantics. However, these tasks do not fully benefit from PLMs since meaning representations are not explicitly included in the pre-training stage. We introduce multilingual pre-trained language-meaning models based on Discourse Representation Structures (DRSs), including meaning representations besides natural language texts in the same model, and design a new strategy to reduce the gap between the pre-training and fine-tuning objectives. Since DRSs are language neutral, cross-lingual transfer learning is adopted to further improve the performance of non-English tasks. Automatic evaluation results show that our approach achieves the best performance on both the multilingual DRS parsing and DRS-to-text generation tasks. Correlation analysis between automatic metrics and human judgements on the generation task further validates the effectiveness of our model. Human inspection reveals that out-of-vocabulary tokens are the main cause of erroneous results.
comment: Accepted by ACL2023 findings
☆ Multilingual Multi-Figurative Language Detection ACL 2023
Figures of speech help people express abstract concepts and evoke stronger emotions than literal expressions, thereby making texts more creative and engaging. Due to its pervasive and fundamental character, figurative language understanding has been addressed in Natural Language Processing, but it's highly understudied in a multilingual setting and when considering more than one figure of speech at the same time. To bridge this gap, we introduce multilingual multi-figurative language modelling, and provide a benchmark for sentence-level figurative language detection, covering three common figures of speech and seven languages. Specifically, we develop a framework for figurative language detection based on template-based prompt learning. In so doing, we unify multiple detection tasks that are interrelated across multiple figures of speech and languages, without requiring task- or language-specific modules. Experimental results show that our framework outperforms several strong baselines and may serve as a blueprint for the joint modelling of other interrelated tasks.
comment: Accepted to ACL 2023 (Findings)
☆ MuseCoco: Generating Symbolic Music from Text
Generating music from text descriptions is a user-friendly mode since the text is a relatively easy interface for user engagement. While some approaches utilize texts to control music audio generation, editing musical elements in generated audio is challenging for users. In contrast, symbolic music offers ease of editing, making it more accessible for users to manipulate specific musical elements. In this paper, we propose MuseCoco, which generates symbolic music from text descriptions with musical attributes as the bridge to break down the task into text-to-attribute understanding and attribute-to-music generation stages. MuseCoCo stands for Music Composition Copilot that empowers musicians to generate music directly from given text descriptions, offering a significant improvement in efficiency compared to creating music entirely from scratch. The system has two main advantages: Firstly, it is data efficient. In the attribute-to-music generation stage, the attributes can be directly extracted from music sequences, making the model training self-supervised. In the text-to-attribute understanding stage, the text is synthesized and refined by ChatGPT based on the defined attribute templates. Secondly, the system can achieve precise control with specific attributes in text descriptions and offers multiple control options through attribute-conditioned or text-conditioned approaches. MuseCoco outperforms baseline systems in terms of musicality, controllability, and overall score by at least 1.27, 1.08, and 1.32 respectively. Besides, there is a notable enhancement of about 20% in objective control accuracy. In addition, we have developed a robust large-scale model with 1.2 billion parameters, showcasing exceptional controllability and musicality.
☆ MERT: Acoustic Music Understanding Model with Large-Scale Self-supervised Training
Self-supervised learning (SSL) has recently emerged as a promising paradigm for training generalisable models on large-scale data in the fields of vision, text, and speech. Although SSL has been proven effective in speech and audio, its application to music audio has yet to be thoroughly explored. This is primarily due to the distinctive challenges associated with modelling musical knowledge, particularly its tonal and pitched characteristics of music. To address this research gap, we propose an acoustic Music undERstanding model with large-scale self-supervised Training (MERT), which incorporates teacher models to provide pseudo labels in the masked language modelling (MLM) style acoustic pre-training. In our exploration, we identified a superior combination of teacher models, which outperforms conventional speech and audio approaches in terms of performance. This combination includes an acoustic teacher based on Residual Vector Quantization - Variational AutoEncoder (RVQ-VAE) and a musical teacher based on the Constant-Q Transform (CQT). These teachers effectively guide our student model, a BERT-style transformer encoder, to better model music audio. In addition, we introduce an in-batch noise mixture augmentation to enhance the representation robustness. Furthermore, we explore a wide range of settings to overcome the instability in acoustic language model pre-training, which allows our designed paradigm to scale from 95M to 330M parameters. Experimental results indicate that our model can generalise and perform well on 14 music understanding tasks and attains state-of-the-art (SOTA) overall scores. The code and models are online: https://github.com/yizhilll/MERT.
☆ ManagerTower: Aggregating the Insights of Uni-Modal Experts for Vision-Language Representation Learning ACL 2023
Two-Tower Vision-Language (VL) models have shown promising improvements on various downstream VL tasks. Although the most advanced work improves performance by building bridges between encoders, it suffers from ineffective layer-by-layer utilization of uni-modal representations and cannot flexibly exploit different levels of uni-modal semantic knowledge. In this work, we propose ManagerTower, a novel VL model architecture that gathers and combines the insights of pre-trained uni-modal experts at different levels. The managers introduced in each cross-modal layer can adaptively aggregate uni-modal semantic knowledge to facilitate more comprehensive cross-modal alignment and fusion. ManagerTower outperforms previous strong baselines both with and without Vision-Language Pre-training (VLP). With only 4M VLP data, ManagerTower achieves superior performances on various downstream VL tasks, especially 79.15% accuracy on VQAv2 Test-Std, 86.56% IR@1 and 95.64% TR@1 on Flickr30K. Code and checkpoints are available at https://github.com/LooperXX/ManagerTower.
comment: Accepted by ACL 2023 Main Conference, Oral
☆ MetaXLR -- Mixed Language Meta Representation Transformation for Low-resource Cross-lingual Learning based on Multi-Armed Bandit
Transfer learning for extremely low resource languages is a challenging task as there is no large scale monolingual corpora for pre training or sufficient annotated data for fine tuning. We follow the work of MetaXL which suggests using meta learning for transfer learning from a single source language to an extremely low resource one. We propose an enhanced approach which uses multiple source languages chosen in a data driven manner. In addition, we introduce a sample selection strategy for utilizing the languages in training by using a multi armed bandit algorithm. Using both of these improvements we managed to achieve state of the art results on the NER task for the extremely low resource languages while using the same amount of data, making the representations better generalized. Also, due to the method ability to use multiple languages it allows the framework to use much larger amounts of data, while still having superior results over the former MetaXL method even with the same amounts of data.
♻ ☆ Scalable Performance Analysis for Vision-Language Models
Joint vision-language models have shown great performance over a diverse set of tasks. However, little is known about their limitations, as the high dimensional space learned by these models makes it difficult to identify semantic errors. Recent work has addressed this problem by designing highly controlled probing task benchmarks. Our paper introduces a more scalable solution that relies on already annotated benchmarks. Our method consists of extracting a large set of diverse features from a vision-language benchmark and measuring their correlation with the output of the target model. We confirm previous findings that CLIP behaves like a bag of words model and performs better with nouns and verbs; we also uncover novel insights such as CLIP getting confused by concrete words. Our framework is available at https://github.com/MichiganNLP/Scalable-VLM-Probing and can be used with other multimodal models and benchmarks.
comment: Camera-ready version for *SEM 2023
♻ ☆ RARR: Researching and Revising What Language Models Say, Using Language Models ACL 2023
Language models (LMs) now excel at many tasks such as few-shot learning, question answering, reasoning, and dialog. However, they sometimes generate unsupported or misleading content. A user cannot easily determine whether their outputs are trustworthy or not, because most LMs do not have any built-in mechanism for attribution to external evidence. To enable attribution while still preserving all the powerful advantages of recent generation models, we propose RARR (Retrofit Attribution using Research and Revision), a system that 1) automatically finds attribution for the output of any text generation model and 2) post-edits the output to fix unsupported content while preserving the original output as much as possible. When applied to the output of several state-of-the-art LMs on a diverse set of generation tasks, we find that RARR significantly improves attribution while otherwise preserving the original input to a much greater degree than previously explored edit models. Furthermore, the implementation of RARR requires only a handful of training examples, a large language model, and standard web search.
comment: ACL 2023
♻ ☆ Generic Temporal Reasoning with Differential Analysis and Explanation
Temporal reasoning is the task of predicting temporal relations of event pairs. While temporal reasoning models can perform reasonably well on in-domain benchmarks, we have little idea of these systems' generalizability due to existing datasets' limitations. In this work, we introduce a novel task named TODAY that bridges this gap with temporal differential analysis, which as the name suggests, evaluates whether systems can correctly understand the effect of incremental changes. Specifically, TODAY introduces slight contextual changes for given event pairs, and systems are asked to tell how this subtle contextual change would affect relevant temporal relation distributions. To facilitate learning, TODAY also annotates human explanations. We show that existing models, including GPT-3.5, drop to random guessing on TODAY, suggesting that they heavily rely on spurious information rather than proper reasoning for temporal predictions. On the other hand, we show that TODAY's supervision style and explanation annotations can be used in joint learning, encouraging models to use more appropriate signals during training and thus outperform across several benchmarks. TODAY can also be used to train models to solicit incidental supervision from noisy sources such as GPT-3.5, thus moving us more toward the goal of generic temporal reasoning systems.
♻ ☆ Pythia: A Suite for Analyzing Large Language Models Across Training and Scaling
How do large language models (LLMs) develop and evolve over the course of training? How do these patterns change as models scale? To answer these questions, we introduce \textit{Pythia}, a suite of 16 LLMs all trained on public data seen in the exact same order and ranging in size from 70M to 12B parameters. We provide public access to 154 checkpoints for each one of the 16 models, alongside tools to download and reconstruct their exact training dataloaders for further study. We intend \textit{Pythia} to facilitate research in many areas, and we present several case studies including novel results in memorization, term frequency effects on few-shot performance, and reducing gender bias. We demonstrate that this highly controlled setup can be used to yield novel insights toward LLMs and their training dynamics. Trained models, analysis code, training code, and training data can be found at \url{https://github.com/EleutherAI/pythia}.
comment: Code at https://github.com/EleutherAI/pythia
♻ ☆ Accuracy on the Curve: On the Nonlinear Correlation of ML Performance Between Data Subpopulations ICML 2023
Understanding the performance of machine learning (ML) models across diverse data distributions is critically important for reliable applications. Despite recent empirical studies positing a near-perfect linear correlation between in-distribution (ID) and out-of-distribution (OOD) accuracies, we empirically demonstrate that this correlation is more nuanced under subpopulation shifts. Through rigorous experimentation and analysis across a variety of datasets, models, and training epochs, we demonstrate that OOD performance often has a nonlinear correlation with ID performance in subpopulation shifts. Our findings, which contrast previous studies that have posited a linear correlation in model performance during distribution shifts, reveal a "moon shape" correlation (parabolic uptrend curve) between the test performance on the majority subpopulation and the minority subpopulation. This non-trivial nonlinear correlation holds across model architectures, hyperparameters, training durations, and the imbalance between subpopulations. Furthermore, we found that the nonlinearity of this "moon shape" is causally influenced by the degree of spurious correlations in the training data. Our controlled experiments show that stronger spurious correlation in the training data creates more nonlinear performance correlation. We provide complementary experimental and theoretical analyses for this phenomenon, and discuss its implications for ML reliability and fairness. Our work highlights the importance of understanding the nonlinear effects of model improvement on performance in different subpopulations, and has the potential to inform the development of more equitable and responsible machine learning models.
comment: Accepted to the main conference of ICML 2023
♻ ☆ Controlled Text Generation with Hidden Representation Transformations ACL 2023
We propose CHRT (Control Hidden Representation Transformation) - a controlled language generation framework that steers large language models to generate text pertaining to certain attributes (such as toxicity). CHRT gains attribute control by modifying the hidden representation of the base model through learned transformations. We employ a contrastive-learning framework to learn these transformations that can be combined to gain multi-attribute control. The effectiveness of CHRT is experimentally shown by comparing it with seven baselines over three attributes. CHRT outperforms all the baselines in the task of detoxification, positive sentiment steering, and text simplification while minimizing the loss in linguistic qualities. Further, our approach has the lowest inference latency of only 0.01 seconds more than the base model, making it the most suitable for high-performance production environments. We open-source our code and release two novel datasets to further propel controlled language generation research.
comment: Accepted at ACL 2023 as a long paper (Findings)
♻ ☆ Grammar Prompting for Domain-Specific Language Generation with Large Language Models
Large language models (LLMs) can learn to perform a wide range of natural language tasks from just a handful of in-context examples. However, for generating strings from highly structured languages (e.g., semantic parsing to complex domain-specific languages), it is challenging for the LLM to generalize from just a few exemplars. We explore $\textbf{grammar prompting}$ as a simple approach for enabling LLMs to use external knowledge and domain-specific constraints, expressed through a grammar expressed in Backus--Naur Form (BNF), during in-context learning. Grammar prompting augments each demonstration example with a specialized grammar that is minimally sufficient for generating the particular output example, where the specialized grammar is a subset of the full DSL grammar. For inference, the LLM first predicts a BNF grammar given a test input, and then generates the output according to the rules of the grammar. Experiments demonstrate that grammar prompting can enable LLMs to perform competitively on a diverse set of DSL generation tasks, including semantic parsing (SMCalFlow, Overnight, GeoQuery), PDDL planning, and even molecule generation (SMILES).
comment: 24 pages with appendix
♻ ☆ Decepticons: Corrupted Transformers Breach Privacy in Federated Learning for Language Models ICLR 2023
A central tenet of Federated learning (FL), which trains models without centralizing user data, is privacy. However, previous work has shown that the gradient updates used in FL can leak user information. While the most industrial uses of FL are for text applications (e.g. keystroke prediction), nearly all attacks on FL privacy have focused on simple image classifiers. We propose a novel attack that reveals private user text by deploying malicious parameter vectors, and which succeeds even with mini-batches, multiple users, and long sequences. Unlike previous attacks on FL, the attack exploits characteristics of both the Transformer architecture and the token embedding, separately extracting tokens and positional embeddings to retrieve high-fidelity text. This work suggests that FL on text, which has historically been resistant to privacy attacks, is far more vulnerable than previously thought.
comment: First two authors contributed equally. Order chosen by coin flip. Published at ICLR 2023. Implementation available at github.com/JonasGeiping/breaching
♻ ☆ Attend-and-Excite: Attention-Based Semantic Guidance for Text-to-Image Diffusion Models SIGGRAPH 2023
Recent text-to-image generative models have demonstrated an unparalleled ability to generate diverse and creative imagery guided by a target text prompt. While revolutionary, current state-of-the-art diffusion models may still fail in generating images that fully convey the semantics in the given text prompt. We analyze the publicly available Stable Diffusion model and assess the existence of catastrophic neglect, where the model fails to generate one or more of the subjects from the input prompt. Moreover, we find that in some cases the model also fails to correctly bind attributes (e.g., colors) to their corresponding subjects. To help mitigate these failure cases, we introduce the concept of Generative Semantic Nursing (GSN), where we seek to intervene in the generative process on the fly during inference time to improve the faithfulness of the generated images. Using an attention-based formulation of GSN, dubbed Attend-and-Excite, we guide the model to refine the cross-attention units to attend to all subject tokens in the text prompt and strengthen - or excite - their activations, encouraging the model to generate all subjects described in the text prompt. We compare our approach to alternative approaches and demonstrate that it conveys the desired concepts more faithfully across a range of text prompts.
comment: Accepted to SIGGRAPH 2023; Project page available at https://yuval-alaluf.github.io/Attend-and-Excite/
♻ ☆ ILLUME: Rationalizing Vision-Language Models through Human Interactions ICML
Bootstrapping from pre-trained language models has been proven to be an efficient approach for building vision-language models (VLM) for tasks such as image captioning or visual question answering. However, outputs of these models rarely align with user's rationales for specific answers. In order to improve this alignment and reinforce commonsense reasons, we propose a tuning paradigm based on human interactions with machine-generated data. Our ILLUME executes the following loop: Given an image-question-answer prompt, the VLM samples multiple candidate rationales, and a human critic provides feedback via preference selection, used for fine-tuning. This loop increases the training data and gradually carves out the VLM's rationalization capabilities that are aligned with human intent. Our exhaustive experiments demonstrate that ILLUME is competitive with standard supervised finetuning while using significantly fewer training data and only requiring minimal feedback.
comment: Proceedings of the 40th International Conference on Machine Learning (ICML), 2023
♻ ☆ Multi-Document Summarization with Centroid-Based Pretraining ACL 2023
In Multi-Document Summarization (MDS), the input can be modeled as a set of documents, and the output is its summary. In this paper, we focus on pretraining objectives for MDS. Specifically, we introduce a novel pretraining objective, which involves selecting the ROUGE-based centroid of each document cluster as a proxy for its summary. Our objective thus does not require human written summaries and can be utilized for pretraining on a dataset consisting solely of document sets. Through zero-shot, few-shot, and fully supervised experiments on multiple MDS datasets, we show that our model Centrum is better or comparable to a state-of-the-art model. We make the pretrained and fine-tuned models freely available to the research community https://github.com/ratishsp/centrum.
comment: ACL 2023 camera-ready
♻ ☆ Query-Utterance Attention with Joint modeling for Query-Focused Meeting Summarization ICASSP 2023
Query-focused meeting summarization (QFMS) aims to generate summaries from meeting transcripts in response to a given query. Previous works typically concatenate the query with meeting transcripts and implicitly model the query relevance only at the token level with attention mechanism. However, due to the dilution of key query-relevant information caused by long meeting transcripts, the original transformer-based model is insufficient to highlight the key parts related to the query. In this paper, we propose a query-aware framework with joint modeling token and utterance based on Query-Utterance Attention. It calculates the utterance-level relevance to the query with a dense retrieval module. Then both token-level query relevance and utterance-level query relevance are combined and incorporated into the generation process with attention mechanism explicitly. We show that the query relevance of different granularities contributes to generating a summary more related to the query. Experimental results on the QMSum dataset show that the proposed model achieves new state-of-the-art performance.
comment: Accepted by ICASSP 2023
♻ ☆ MT4SSL: Boosting Self-Supervised Speech Representation Learning by Integrating Multiple Targets
In this paper, we provide a new perspective on self-supervised speech models from how the training targets are obtained. We generalize the targets extractor into Offline Targets Extractor (Off-TE) and Online Targets Extractor (On-TE). Based on this, we propose a new multi-tasking learning framework for self-supervised learning, MT4SSL, which stands for Boosting Self-Supervised Speech Representation Learning by Integrating Multiple Targets. MT4SSL uses the K-means algorithm as an Off-TE and a teacher network without gradients as an On-TE, respectively. Our model outperforms previous SSL methods by nontrivial margins on the LibriSpeech benchmark, and is comparable to or even better than the best-performing models with fewer data. Furthermore, we find that using both Off-TE and On-TE results in better convergence in the pre-training phase. With both effectiveness and efficiency, we think doing multi-task learning on self-supervised speech models from our perspective is a promising trend.
comment: Accepted to Interspeech 2023. Code available at: https://github.com/ddlBoJack/MT4SSL
♻ ☆ AmbiFC: Fact-Checking Ambiguous Claims with Evidence
Automated fact-checking systems in real-world scenarios must compare claims with retrieved evidence to predict the veracity. The retrieved evidence may not unambiguously support or refute the claim and yield diverse valid interpretations. Existing fact-checking datasets necessitate that models predict a single veracity label for each claim and lack the ability to manage such ambiguity. We present AmbiFC, a large-scale fact-checking dataset with realistic claims derived from real-world information needs. Our dataset contains fine-grained evidence annotations of passages from complete Wikipedia pages. We thoroughly analyze disagreements arising from ambiguous claims in AmbiFC, observing a strong correlation of annotator disagreement with their self-assessment and expert-annotated linguistic phenomena. We introduce the task of evidence-based fact-checking for ambiguous claims with soft labels, and compare three methodologies incorporating annotation signals with a single-label classification approach. We find that a pipeline with annotation distillation for sentence-level evidence selection and veracity prediction yields the best performance. Models trained on ambiguous instances exhibit improved performance dealing with the identified linguistic categories, and acquire an understanding of nuanced differences among evidence sentences associated with diverse veracity interpretations.
comment: Code and Data https://github.com/CambridgeNLIP/verification-real-world-info-needs
♻ ☆ Variational Open-Domain Question Answering ICML 2023
Retrieval-augmented models have proven to be effective in natural language processing tasks, yet there remains a lack of research on their optimization using variational inference. We introduce the Variational Open-Domain (VOD) framework for end-to-end training and evaluation of retrieval-augmented models, focusing on open-domain question answering and language modelling. The VOD objective, a self-normalized estimate of the R\'enyi variational bound, approximates the task marginal likelihood and is evaluated under samples drawn from an auxiliary sampling distribution (cached retriever and/or approximate posterior). It remains tractable, even for retriever distributions defined on large corpora. We demonstrate VOD's versatility by training reader-retriever BERT-sized models on multiple-choice medical exam questions. On the MedMCQA dataset, we outperform the domain-tuned Med-PaLM by +5.3% despite using 2.500$\times$ fewer parameters. Our retrieval-augmented BioLinkBERT model scored 62.9% on the MedMCQA and 55.0% on the MedQA-USMLE. Last, we show the effectiveness of our learned retriever component in the context of medical semantic search.
comment: 28 pages, 5 figures. Accepted at ICML 2023
♻ ☆ The Curse of Recursion: Training on Generated Data Makes Models Forget
Stable Diffusion revolutionised image creation from descriptive text. GPT-2, GPT-3(.5) and GPT-4 demonstrated astonishing performance across a variety of language tasks. ChatGPT introduced such language models to the general public. It is now clear that large language models (LLMs) are here to stay, and will bring about drastic change in the whole ecosystem of online text and images. In this paper we consider what the future might hold. What will happen to GPT-{n} once LLMs contribute much of the language found online? We find that use of model-generated content in training causes irreversible defects in the resulting models, where tails of the original content distribution disappear. We refer to this effect as Model Collapse and show that it can occur in Variational Autoencoders, Gaussian Mixture Models and LLMs. We build theoretical intuition behind the phenomenon and portray its ubiquity amongst all learned generative models. We demonstrate that it has to be taken seriously if we are to sustain the benefits of training from large-scale data scraped from the web. Indeed, the value of data collected about genuine human interactions with systems will be increasingly valuable in the presence of content generated by LLMs in data crawled from the Internet.
♻ ☆ UPop: Unified and Progressive Pruning for Compressing Vision-Language Transformers ICML 2023
Real-world data contains a vast amount of multimodal information, among which vision and language are the two most representative modalities. Moreover, increasingly heavier models, \textit{e}.\textit{g}., Transformers, have attracted the attention of researchers to model compression. However, how to compress multimodal models, especially vison-language Transformers, is still under-explored. This paper proposes the \textbf{U}nified and \textbf{P}r\textbf{o}gressive \textbf{P}runing (\textbf{\emph{UPop}}) as a universal vison-language Transformer compression framework, which incorporates 1) unifiedly searching multimodal subnets in a continuous optimization space from the original model, which enables automatic assignment of pruning ratios among compressible modalities and structures; 2) progressively searching and retraining the subnet, which maintains convergence between the search and retrain to attain higher compression ratios. Experiments on various tasks, datasets, and model architectures demonstrate the effectiveness and versatility of the proposed UPop framework. The code is available at https://github.com/sdc17/UPop.
comment: ICML 2023
♻ ☆ On Bias and Fairness in NLP: How to have a fairer text classification?
In this paper, we provide a holistic analysis of the different sources of bias, Upstream, Sample and Overampflication biases, in NLP models. We investigate how they impact the fairness of the task of text classification. We also investigate the impact of removing these biases using different debiasing techniques on the fairness of text classification. We found that overamplification bias is the most impactful bias on the fairness of text classification. And that removing overamplification bias by fine-tuning the LM models on a dataset with balanced representations of the different identity groups leads to fairer text classification models. Finally, we build on our findings and introduce practical guidelines on how to have a fairer text classification model.
comment: 10 pages
♻ ☆ Transformers learn in-context by gradient descent
At present, the mechanisms of in-context learning in Transformers are not well understood and remain mostly an intuition. In this paper, we suggest that training Transformers on auto-regressive objectives is closely related to gradient-based meta-learning formulations. We start by providing a simple weight construction that shows the equivalence of data transformations induced by 1) a single linear self-attention layer and by 2) gradient-descent (GD) on a regression loss. Motivated by that construction, we show empirically that when training self-attention-only Transformers on simple regression tasks either the models learned by GD and Transformers show great similarity or, remarkably, the weights found by optimization match the construction. Thus we show how trained Transformers become mesa-optimizers i.e. learn models by gradient descent in their forward pass. This allows us, at least in the domain of regression problems, to mechanistically understand the inner workings of in-context learning in optimized Transformers. Building on this insight, we furthermore identify how Transformers surpass the performance of plain gradient descent by learning an iterative curvature correction and learn linear models on deep data representations to solve non-linear regression tasks. Finally, we discuss intriguing parallels to a mechanism identified to be crucial for in-context learning termed induction-head (Olsson et al., 2022) and show how it could be understood as a specific case of in-context learning by gradient descent learning within Transformers. Code to reproduce the experiments can be found at https://github.com/google-research/self-organising-systems/tree/master/transformers_learn_icl_by_gd .
♻ ☆ On the Hidden Mystery of OCR in Large Multimodal Models
Large models have recently played a dominant role in natural language processing and multimodal vision-language learning. It remains less explored about their efficacy in text-related visual tasks. We conducted a comprehensive study of existing publicly available multimodal models, evaluating their performance in text recognition (document text, artistic text, handwritten text, scene text), text-based visual question answering (document text, scene text, and bilingual text), key information extraction (receipts, documents, and nutrition facts) and handwritten mathematical expression recognition. Our findings reveal strengths and weaknesses in these models, which primarily rely on semantic understanding for word recognition and exhibit inferior perception of individual character shapes. They also display indifference towards text length and have limited capabilities in detecting fine-grained features in images. Consequently, these results demonstrate that even the current most powerful large multimodal models cannot match domain-specific methods in traditional text tasks and face greater challenges in more complex tasks. Most importantly, the baseline results showcased in this study could provide a foundational framework for the conception and assessment of innovative strategies targeted at enhancing zero-shot multimodal techniques. Evaluation pipeline will be available at https://github.com/Yuliang-Liu/MultimodalOCR.
♻ ☆ Denoising Bottleneck with Mutual Information Maximization for Video Multimodal Fusion ACL2023
Video multimodal fusion aims to integrate multimodal signals in videos, such as visual, audio and text, to make a complementary prediction with multiple modalities contents. However, unlike other image-text multimodal tasks, video has longer multimodal sequences with more redundancy and noise in both visual and audio modalities. Prior denoising methods like forget gate are coarse in the granularity of noise filtering. They often suppress the redundant and noisy information at the risk of losing critical information. Therefore, we propose a denoising bottleneck fusion (DBF) model for fine-grained video multimodal fusion. On the one hand, we employ a bottleneck mechanism to filter out noise and redundancy with a restrained receptive field. On the other hand, we use a mutual information maximization module to regulate the filter-out module to preserve key information within different modalities. Our DBF model achieves significant improvement over current state-of-the-art baselines on multiple benchmarks covering multimodal sentiment analysis and multimodal summarization tasks. It proves that our model can effectively capture salient features from noisy and redundant video, audio, and text inputs. The code for this paper is publicly available at https://github.com/WSXRHFG/DBF.
comment: Accept at ACL2023
♻ ☆ Cross-Lingual Retrieval Augmented Prompt for Low-Resource Languages ACL 2023
Multilingual Pretrained Language Models (MPLMs) have shown their strong multilinguality in recent empirical cross-lingual transfer studies. In this paper, we propose the Prompts Augmented by Retrieval Crosslingually (PARC) pipeline to improve the zero-shot performance on low-resource languages (LRLs) by augmenting the context with semantically similar sentences retrieved from a high-resource language (HRL) as prompts. PARC improves the zero-shot performance on three downstream tasks (binary sentiment classification, topic categorization and natural language inference) with multilingual parallel test sets across 10 LRLs covering 6 language families in both unlabeled settings (+5.1%) and labeled settings (+16.3%). PARC-labeled also outperforms the finetuning baseline by 3.7%. We find a significant positive correlation between cross-lingual transfer performance on one side, and the similarity between the high- and low-resource languages as well as the amount of low-resource pretraining data on the other side. A robustness analysis suggests that PARC has the potential to achieve even stronger performance with more powerful MPLMs.
comment: Accepted to Findings of ACL 2023
♻ ☆ Exploring Partial Knowledge Base Inference in Biomedical Entity Linking ACL
Biomedical entity linking (EL) consists of named entity recognition (NER) and named entity disambiguation (NED). EL models are trained on corpora labeled by a predefined KB. However, it is a common scenario that only entities within a subset of the KB are precious to stakeholders. We name this scenario partial knowledge base inference: training an EL model with one KB and inferring on the part of it without further training. In this work, we give a detailed definition and evaluation procedures for this practically valuable but significantly understudied scenario and evaluate methods from three representative EL paradigms. We construct partial KB inference benchmarks and witness a catastrophic degradation in EL performance due to dramatically precision drop. Our findings reveal these EL paradigms can not correctly handle unlinkable mentions (NIL), so they are not robust to partial KB inference. We also propose two simple-and-effective redemption methods to combat the NIL issue with little computational overhead.
comment: Accepted by ACL-BioNLP 2023. The first two authors are contributed equally
♻ ☆ A Framework For Refining Text Classification and Object Recognition from Academic Articles
With the widespread use of the internet, it has become increasingly crucial to extract specific information from vast amounts of academic articles efficiently. Data mining techniques are generally employed to solve this issue. However, data mining for academic articles is challenging since it requires automatically extracting specific patterns in complex and unstructured layout documents. Current data mining methods for academic articles employ rule-based(RB) or machine learning(ML) approaches. However, using rule-based methods incurs a high coding cost for complex typesetting articles. On the other hand, simply using machine learning methods requires annotation work for complex content types within the paper, which can be costly. Furthermore, only using machine learning can lead to cases where patterns easily recognized by rule-based methods are mistakenly extracted. To overcome these issues, from the perspective of analyzing the standard layout and typesetting used in the specified publication, we emphasize implementing specific methods for specific characteristics in academic articles. We have developed a novel Text Block Refinement Framework (TBRF), a machine learning and rule-based scheme hybrid. We used the well-known ACL proceeding articles as experimental data for the validation experiment. The experiment shows that our approach achieved over 95% classification accuracy and 90% detection accuracy for tables and figures.
comment: This paper will be submitted to 'The International Symposium on Innovations in Intelligent SysTems and Applications 2023 (INISTA 2023)'
♻ ☆ DuNST: Dual Noisy Self Training for Semi-Supervised Controllable Text Generation
Self-training (ST) has prospered again in language understanding by augmenting the fine-tuning of pre-trained language models when labeled data is insufficient. However, it remains challenging to incorporate ST into attribute-controllable language generation. Augmented by only self-generated pseudo text, generation models over-emphasize exploitation of the previously learned space, suffering from a constrained generalization boundary. We revisit ST and propose a novel method, DuNST to alleviate this problem. DuNST jointly models text generation and classification with a shared Variational AutoEncoder and corrupts the generated pseudo text by two kinds of flexible noise to disturb the space. In this way, our model could construct and utilize both pseudo text from given labels and pseudo labels from available unlabeled text, which are gradually refined during the ST process. We theoretically demonstrate that DuNST can be regarded as enhancing exploration towards the potential real text space, providing a guarantee of improved performance. Experiments on three controllable generation tasks show that DuNST could significantly boost control accuracy while maintaining comparable generation fluency and diversity against several strong baselines.
♻ ☆ Task-Optimized Adapters for an End-to-End Task-Oriented Dialogue System ACL2023
Task-Oriented Dialogue (TOD) systems are designed to carry out specific tasks by tracking dialogue states and generating appropriate responses to help users achieve defined goals. Recently, end-to-end dialogue models pre-trained based on large datasets have shown promising performance in the conversational system. However, they share the same parameters to train tasks of the dialogue system (NLU, DST, NLG), so debugging each task is challenging. Also, they require a lot of effort to fine-tune large parameters to create a task-oriented chatbot, making it difficult for non-experts to handle. Therefore, we intend to train relatively lightweight and fast models compared to PLM. In this paper, we propose an End-to-end TOD system with Task-Optimized Adapters which learn independently per task, adding only small number of parameters after fixed layers of pre-trained network. We also enhance the performance of the DST and NLG modules through reinforcement learning, overcoming the learning curve that has lacked at the adapter learning and enabling the natural and consistent response generation that is appropriate for the goal. Our method is a model-agnostic approach and does not require prompt-tuning as only input data without a prompt. As results of the experiment, our method shows competitive performance on the MultiWOZ benchmark compared to the existing end-to-end models. In particular, we attain state-of-the-art performance on the DST task of 2.2 dataset.
comment: Accepted to Findings of ACL2023
♻ ☆ Practical PCG Through Large Language Models
Large Language Models (LLMs) have proven to be useful tools in various domains outside of the field of their inception, which was natural language processing. In this study, we provide practical directions on how to use LLMs to generate 2D-game rooms for an under-development game, named Metavoidal. Our technique can harness the power of GPT-3 by Human-in-the-loop fine-tuning which allows our method to create 37% Playable-Novel levels from as scarce data as only 60 hand-designed rooms under a scenario of the non-trivial game, with respect to (Procedural Content Generation) PCG, that has a good amount of local and global constraints.
♻ ☆ Continual Contrastive Finetuning Improves Low-Resource Relation Extraction ACL 2023
Relation extraction (RE), which has relied on structurally annotated corpora for model training, has been particularly challenging in low-resource scenarios and domains. Recent literature has tackled low-resource RE by self-supervised learning, where the solution involves pretraining the entity pair embedding by RE-based objective and finetuning on labeled data by classification-based objective. However, a critical challenge to this approach is the gap in objectives, which prevents the RE model from fully utilizing the knowledge in pretrained representations. In this paper, we aim at bridging the gap and propose to pretrain and finetune the RE model using consistent objectives of contrastive learning. Since in this kind of representation learning paradigm, one relation may easily form multiple clusters in the representation space, we further propose a multi-center contrastive loss that allows one relation to form multiple clusters to better align with pretraining. Experiments on two document-level RE datasets, BioRED and Re-DocRED, demonstrate the effectiveness of our method. Particularly, when using 1% end-task training data, our method outperforms PLM-based RE classifier by 10.5% and 6.1% on the two datasets, respectively.
comment: ACL 2023
♻ ☆ QUEST: A Retrieval Dataset of Entity-Seeking Queries with Implicit Set Operations ACL 2023
Formulating selective information needs results in queries that implicitly specify set operations, such as intersection, union, and difference. For instance, one might search for "shorebirds that are not sandpipers" or "science-fiction films shot in England". To study the ability of retrieval systems to meet such information needs, we construct QUEST, a dataset of 3357 natural language queries with implicit set operations, that map to a set of entities corresponding to Wikipedia documents. The dataset challenges models to match multiple constraints mentioned in queries with corresponding evidence in documents and correctly perform various set operations. The dataset is constructed semi-automatically using Wikipedia category names. Queries are automatically composed from individual categories, then paraphrased and further validated for naturalness and fluency by crowdworkers. Crowdworkers also assess the relevance of entities based on their documents and highlight attribution of query constraints to spans of document text. We analyze several modern retrieval systems, finding that they often struggle on such queries. Queries involving negation and conjunction are particularly challenging and systems are further challenged with combinations of these operations.
comment: ACL 2023; Dataset available at https://github.com/google-research/language/tree/master/language/quest
♻ ☆ Matching Exemplar as Next Sentence Prediction (MeNSP): Zero-shot Prompt Learning for Automatic Scoring in Science Education
Developing models to automatically score students' written responses to science problems is critical for science education. However, collecting and labeling sufficient student responses for training models is time and cost-consuming. Recent studies suggest that pre-trained language models can be adapted to downstream tasks without fine-tuning with prompts. However, no research has employed such a prompt approach in science education. As student responses are presented with natural language, aligning the scoring procedure as the next sentence prediction task using prompts can skip the costly fine-tuning stage. In this study, we developed a zero-shot approach to automatically score student responses via Matching Exemplars as Next Sentence Prediction (MeNSP). This approach employs no training samples. We first apply MeNSP in scoring three assessment tasks of scientific argumentation and found machine-human scoring agreements, Cohen's Kappa ranges from 0.30 to 0.57, and F1 score ranges from 0.54 to 0.81. To improve the performance, we extend our research to the few-shots setting, either randomly selecting labeled student responses or manually constructing responses to fine-tune the models. We find that one task's performance is improved with more samples, Cohen's Kappa from 0.30 to 0.38, and F1 score from 0.54 to 0.59; for the two others, scoring performance is not improved. We also find that randomly selected few-shots perform better than the human expert-crafted approach. This study suggests that MeNSP can yield referable automatic scoring for student responses while significantly reducing the cost of model training. This method can benefit low-stakes classroom assessment practices in science education. Future research should further explore the applicability of the MeNSP in different types of assessment tasks in science education and improve the model performance.
comment: 10 page + 1 figure
♻ ☆ Do GPTs Produce Less Literal Translations? ACL 2023
Large Language Models (LLMs) such as GPT-3 have emerged as general-purpose language models capable of addressing many natural language generation or understanding tasks. On the task of Machine Translation (MT), multiple works have investigated few-shot prompting mechanisms to elicit better translations from LLMs. However, there has been relatively little investigation on how such translations differ qualitatively from the translations generated by standard Neural Machine Translation (NMT) models. In this work, we investigate these differences in terms of the literalness of translations produced by the two systems. Using literalness measures involving word alignment and monotonicity, we find that translations out of English (E-X) from GPTs tend to be less literal, while exhibiting similar or better scores on MT quality metrics. We demonstrate that this finding is borne out in human evaluations as well. We then show that these differences are especially pronounced when translating sentences that contain idiomatic expressions.
comment: ACL 2023
♻ ☆ Cross Encoding as Augmentation: Towards Effective Educational Text Classification ACL2023
Text classification in education, usually called auto-tagging, is the automated process of assigning relevant tags to educational content, such as questions and textbooks. However, auto-tagging suffers from a data scarcity problem, which stems from two major challenges: 1) it possesses a large tag space and 2) it is multi-label. Though a retrieval approach is reportedly good at low-resource scenarios, there have been fewer efforts to directly address the data scarcity problem. To mitigate these issues, here we propose a novel retrieval approach CEAA that provides effective learning in educational text classification. Our main contributions are as follows: 1) we leverage transfer learning from question-answering datasets, and 2) we propose a simple but effective data augmentation method introducing cross-encoder style texts to a bi-encoder architecture for more efficient inference. An extensive set of experiments shows that our proposed method is effective in multi-label scenarios and low-resource tags compared to state-of-the-art models.
comment: Accepted to Findings of ACL2023
♻ ☆ Synthetic Pre-Training Tasks for Neural Machine Translation ACL2023
Pre-training models with large crawled corpora can lead to issues such as toxicity and bias, as well as copyright and privacy concerns. A promising way of alleviating such concerns is to conduct pre-training with synthetic tasks and data, since no real-world information is ingested by the model. Our goal in this paper is to understand the factors that contribute to the effectiveness of pre-training models when using synthetic resources, particularly in the context of neural machine translation. We propose several novel approaches to pre-training translation models that involve different levels of lexical and structural knowledge, including: 1) generating obfuscated data from a large parallel corpus 2) concatenating phrase pairs extracted from a small word-aligned corpus, and 3) generating synthetic parallel data without real human language corpora. Our experiments on multiple language pairs reveal that pre-training benefits can be realized even with high levels of obfuscation or purely synthetic parallel data. We hope the findings from our comprehensive empirical analysis will shed light on understanding what matters for NMT pre-training, as well as pave the way for the development of more efficient and less toxic models.
comment: Accepted to ACL2023-Findings. New added Phrase-cat for synthetic pre-training. 17 pages including 5-page appendix
♻ ☆ Comparison of Multilingual Self-Supervised and Weakly-Supervised Speech Pre-Training for Adaptation to Unseen Languages
Recent models such as XLS-R and Whisper have made multilingual speech technologies more accessible by pre-training on audio from around 100 spoken languages each. However, there are thousands of spoken languages worldwide, and adapting to new languages is an important problem. In this work, we aim to understand which model adapts better to languages unseen during pre-training. We fine-tune both models on 13 unseen languages and 18 seen languages. Our results show that the number of hours seen per language and language family during pre-training is predictive of how the models compare, despite the significant differences in the pre-training methods.
comment: Accepted at Interspeech 2023
♻ ☆ cTBLS: Augmenting Large Language Models with Conversational Tables
Optimizing accuracy and performance while eliminating hallucinations of open-domain conversational large language models (LLMs) is an open research challenge. A particularly promising direction is to augment and ground LLMs with information from structured sources. This paper introduces Conversational Tables (cTBLS), a three-step architecture to retrieve and generate dialogue responses grounded on retrieved tabular information. cTBLS uses Transformer encoder embeddings for Dense Table Retrieval and obtains up to 125% relative improvement over the retriever in the previous state-of-the-art system on the HyrbiDialogue dataset. cTBLS then uses a shared process between encoder and decoder models to perform a coarse+fine tabular knowledge (e.g., cell) ranking combined with a GPT-3.5 LLM response generator to yield a 2x relative improvement in ROUGE scores. Finally, human evaluators prefer cTBLs +80% of the time (coherency, fluency) and judge informativeness to be 4x better than the previous state-of-the-art.
♻ ☆ Beyond One-Model-Fits-All: A Survey of Domain Specialization for Large Language Models
Large language models (LLMs) have significantly advanced the field of natural language processing (NLP), providing a highly useful, task-agnostic foundation for a wide range of applications. The great promise of LLMs as general task solvers motivated people to extend their functionality largely beyond just a ``chatbot'', and use it as an assistant or even replacement for domain experts and tools in specific domains such as healthcare, finance, and education. However, directly applying LLMs to solve sophisticated problems in specific domains meets many hurdles, caused by the heterogeneity of domain data, the sophistication of domain knowledge, the uniqueness of domain objectives, and the diversity of the constraints (e.g., various social norms, cultural conformity, religious beliefs, and ethical standards in the domain applications). To fill such a gap, explosively-increase research, and practices have been conducted in very recent years on the domain specialization of LLMs, which, however, calls for a comprehensive and systematic review to better summarizes and guide this promising domain. In this survey paper, first, we propose a systematic taxonomy that categorizes the LLM domain-specialization techniques based on the accessibility to LLMs and summarizes the framework for all the subcategories as well as their relations and differences to each other. We also present a comprehensive taxonomy of critical application domains that can benefit from specialized LLMs, discussing their practical significance and open challenges. Furthermore, we offer insights into the current research status and future trends in this area.
♻ ☆ Reasoning with Language Model Prompting: A Survey ACL 2023
Reasoning, as an essential ability for complex problem-solving, can provide back-end support for various real-world applications, such as medical diagnosis, negotiation, etc. This paper provides a comprehensive survey of cutting-edge research on reasoning with language model prompting. We introduce research works with comparisons and summaries and provide systematic resources to help beginners. We also discuss the potential reasons for emerging such reasoning abilities and highlight future research directions. Resources are available at https://github.com/zjunlp/Prompt4ReasoningPapers (updated periodically).
comment: ACL 2023
♻ ☆ Representation Of Lexical Stylistic Features In Language Models' Embedding Space
The representation space of pretrained Language Models (LMs) encodes rich information about words and their relationships (e.g., similarity, hypernymy, polysemy) as well as abstract semantic notions (e.g., intensity). In this paper, we demonstrate that lexical stylistic notions such as complexity, formality, and figurativeness, can also be identified in this space. We show that it is possible to derive a vector representation for each of these stylistic notions from only a small number of seed pairs. Using these vectors, we can characterize new texts in terms of these dimensions by performing simple calculations in the corresponding embedding space. We conduct experiments on five datasets and find that static embeddings encode these features more accurately at the level of words and phrases, whereas contextualized LMs perform better on sentences. The lower performance of contextualized representations at the word level is partially attributable to the anisotropy of their vector space, which can be corrected to some extent using techniques like standardization.
comment: Accepted at *SEM 2023
♻ ☆ Emergent Linguistic Structures in Neural Networks are Fragile
Large Language Models (LLMs) have been reported to have strong performance on natural language processing tasks. However, performance metrics such as accuracy do not measure the quality of the model in terms of its ability to robustly represent complex linguistic structures. In this paper, focusing on the ability of language models to represent syntax, we propose a framework to assess the consistency and robustness of linguistic representations. To this end, we introduce measures of robustness of neural network models that leverage recent advances in extracting linguistic constructs from LLMs via probing tasks, i.e., simple tasks used to extract meaningful information about a single facet of a language model, such as syntax reconstruction and root identification. Empirically, we study the performance of four LLMs across six different corpora on the proposed robustness measures by analysing their performance and robustness with respect to syntax-preserving perturbations. We provide evidence that context-free representation (e.g., GloVe) are in some cases competitive with context-dependent representations from modern LLMs (e.g., BERT), yet equally brittle to syntax-preserving perturbations. Our key observation is that emergent syntactic representations in neural networks are brittle. We make the code, trained models and logs available to the community as a contribution to the debate about the capabilities of LLMs.
♻ ☆ Almanac: Retrieval-Augmented Language Models for Clinical Medicine
Large-language models have recently demonstrated impressive zero-shot capabilities in a variety of natural language tasks such as summarization, dialogue generation, and question-answering. Despite many promising applications in clinical medicine, adoption of these models in real-world settings has been largely limited by their tendency to generate incorrect and sometimes even toxic statements. In this study, we develop Almanac, a large language model framework augmented with retrieval capabilities for medical guideline and treatment recommendations. Performance on a novel dataset of clinical scenarios (n = 130) evaluated by a panel of 5 board-certified and resident physicians demonstrates significant increases in factuality (mean of 18% at p-value < 0.05) across all specialties, with improvements in completeness and safety. Our results demonstrate the potential for large language models to be effective tools in the clinical decision-making process, while also emphasizing the importance of careful testing and deployment to mitigate their shortcomings.
♻ ☆ Multimodal Speech Recognition for Language-Guided Embodied Agents ISCA
Benchmarks for language-guided embodied agents typically assume text-based instructions, but deployed agents will encounter spoken instructions. While Automatic Speech Recognition (ASR) models can bridge the input gap, erroneous ASR transcripts can hurt the agents' ability to complete tasks. In this work, we propose training a multimodal ASR model to reduce errors in transcribing spoken instructions by considering the accompanying visual context. We train our model on a dataset of spoken instructions, synthesized from the ALFRED task completion dataset, where we simulate acoustic noise by systematically masking spoken words. We find that utilizing visual observations facilitates masked word recovery, with multimodal ASR models recovering up to 30% more masked words than unimodal baselines. We also find that a text-trained embodied agent successfully completes tasks more often by following transcribed instructions from multimodal ASR models. github.com/Cylumn/embodied-multimodal-asr
comment: 5 pages, 5 figures, 24th ISCA Interspeech Conference (INTERSPEECH 2023)
♻ ☆ NLP Reproducibility For All: Understanding Experiences of Beginners ACL 2023
As natural language processing (NLP) has recently seen an unprecedented level of excitement, and more people are eager to enter the field, it is unclear whether current research reproducibility efforts are sufficient for this group of beginners to apply the latest developments. To understand their needs, we conducted a study with 93 students in an introductory NLP course, where students reproduced the results of recent NLP papers. Surprisingly, we find that their programming skill and comprehension of research papers have a limited impact on their effort spent completing the exercise. Instead, we find accessibility efforts by research authors to be the key to success, including complete documentation, better coding practice, and easier access to data files. Going forward, we recommend that NLP researchers pay close attention to these simple aspects of open-sourcing their work, and use insights from beginners' feedback to provide actionable ideas on how to better support them.
comment: ACL 2023 Theme Track
♻ ☆ Exploring Challenges of Deploying BERT-based NLP Models in Resource-Constrained Embedded Devices
BERT-based neural architectures have established themselves as popular state-of-the-art baselines for many downstream NLP tasks. However, these architectures are data-hungry and consume a lot of memory and energy, often hindering their deployment in many real-time, resource-constrained applications. Existing lighter versions of BERT (eg. DistilBERT and TinyBERT) often cannot perform well on complex NLP tasks. More importantly, from a designer's perspective, it is unclear what is the "right" BERT-based architecture to use for a given NLP task that can strike the optimal trade-off between the resources available and the minimum accuracy desired by the end user. System engineers have to spend a lot of time conducting trial-and-error experiments to find a suitable answer to this question. This paper presents an exploratory study of BERT-based models under different resource constraints and accuracy budgets to derive empirical observations about this resource/accuracy trade-offs. Our findings can help designers to make informed choices among alternative BERT-based architectures for embedded systems, thus saving significant development time and effort.
♻ ☆ Large Language Models Are State-of-the-Art Evaluators of Translation Quality
We describe GEMBA, a GPT-based metric for assessment of translation quality, which works both with a reference translation and without. In our evaluation, we focus on zero-shot prompting, comparing four prompt variants in two modes, based on the availability of the reference. We investigate nine versions of GPT models, including ChatGPT and GPT-4. We show that our method for translation quality assessment only works with GPT~3.5 and larger models. Comparing to results from WMT22's Metrics shared task, our method achieves state-of-the-art accuracy in both modes when compared to MQM-based human labels. Our results are valid on the system level for all three WMT22 Metrics shared task language pairs, namely English into German, English into Russian, and Chinese into English. This provides a first glimpse into the usefulness of pre-trained, generative large language models for quality assessment of translations. We publicly release all our code and prompt templates used for the experiments described in this work, as well as all corresponding scoring results, to allow for external validation and reproducibility.
comment: Accepted in EAMT, 10 pages, 8 tables, one figure
♻ ☆ Emergent and Predictable Memorization in Large Language Models
Memorization, or the tendency of large language models (LLMs) to output entire sequences from their training data verbatim, is a key concern for safely deploying language models. In particular, it is vital to minimize a model's memorization of sensitive datapoints such as those containing personal identifiable information (PII). The prevalence of such undesirable memorization can pose issues for model trainers, and may even require discarding an otherwise functional model. We therefore seek to predict which sequences will be memorized before a large model's full train-time by extrapolating the memorization behavior of lower-compute trial runs. We measure memorization of the Pythia model suite and plot scaling laws for forecasting memorization, allowing us to provide equi-compute recommendations to maximize the reliability (recall) of such predictions. We additionally provide further novel discoveries on the distribution of memorization scores across models and data. We release all code and data necessary to reproduce the results in this paper at https://github.com/EleutherAI/pythia
Computer Vision and Pattern Recognition 144
☆ Humans in 4D: Reconstructing and Tracking Humans with Transformers
We present an approach to reconstruct humans and track them over time. At the core of our approach, we propose a fully "transformerized" version of a network for human mesh recovery. This network, HMR 2.0, advances the state of the art and shows the capability to analyze unusual poses that have in the past been difficult to reconstruct from single images. To analyze video, we use 3D reconstructions from HMR 2.0 as input to a tracking system that operates in 3D. This enables us to deal with multiple people and maintain identities through occlusion events. Our complete approach, 4DHumans, achieves state-of-the-art results for tracking people from monocular video. Furthermore, we demonstrate the effectiveness of HMR 2.0 on the downstream task of action recognition, achieving significant improvements over previous pose-based action recognition approaches. Our code and models are available on the project website: https://shubham-goel.github.io/4dhumans/.
comment: Project Webpage: https://shubham-goel.github.io/4dhumans/
☆ Learning Explicit Contact for Implicit Reconstruction of Hand-held Objects from Monocular Images
Reconstructing hand-held objects from monocular RGB images is an appealing yet challenging task. In this task, contacts between hands and objects provide important cues for recovering the 3D geometry of the hand-held objects. Though recent works have employed implicit functions to achieve impressive progress, they ignore formulating contacts in their frameworks, which results in producing less realistic object meshes. In this work, we explore how to model contacts in an explicit way to benefit the implicit reconstruction of hand-held objects. Our method consists of two components: explicit contact prediction and implicit shape reconstruction. In the first part, we propose a new subtask of directly estimating 3D hand-object contacts from a single image. The part-level and vertex-level graph-based transformers are cascaded and jointly learned in a coarse-to-fine manner for more accurate contact probabilities. In the second part, we introduce a novel method to diffuse estimated contact states from the hand mesh surface to nearby 3D space and leverage diffused contact probabilities to construct the implicit neural representation for the manipulated object. Benefiting from estimating the interaction patterns between the hand and the object, our method can reconstruct more realistic object meshes, especially for object parts that are in contact with hands. Extensive experiments on challenging benchmarks show that the proposed method outperforms the current state of the arts by a great margin.
comment: 17 pages, 8 figures
☆ Improving CLIP Training with Language Rewrites
Contrastive Language-Image Pre-training (CLIP) stands as one of the most effective and scalable methods for training transferable vision models using paired image and text data. CLIP models are trained using contrastive loss, which typically relies on data augmentations to prevent overfitting and shortcuts. However, in the CLIP training paradigm, data augmentations are exclusively applied to image inputs, while language inputs remain unchanged throughout the entire training process, limiting the exposure of diverse texts to the same image. In this paper, we introduce Language augmented CLIP (LaCLIP), a simple yet highly effective approach to enhance CLIP training through language rewrites. Leveraging the in-context learning capability of large language models, we rewrite the text descriptions associated with each image. These rewritten texts exhibit diversity in sentence structure and vocabulary while preserving the original key concepts and meanings. During training, LaCLIP randomly selects either the original texts or the rewritten versions as text augmentations for each image. Extensive experiments on CC3M, CC12M, RedCaps and LAION-400M datasets show that CLIP pre-training with language rewrites significantly improves the transfer performance without computation or memory overhead during training. Specifically for ImageNet zero-shot accuracy, LaCLIP outperforms CLIP by 8.2% on CC12M and 2.4% on LAION-400M. Code is available at https://github.com/LijieFan/LaCLIP.
☆ Too Large; Data Reduction for Vision-Language Pre-Training
This paper examines the problems of severe image-text misalignment and high redundancy in the widely-used large-scale Vision-Language Pre-Training (VLP) datasets. To address these issues, we propose an efficient and straightforward Vision-Language learning algorithm called TL;DR, which aims to compress the existing large VLP data into a small, high-quality set. Our approach consists of two major steps. First, a codebook-based encoder-decoder captioner is developed to select representative samples. Second, a new caption is generated to complement the original captions for selected samples, mitigating the text-image misalignment problem while maintaining uniqueness. As the result, TL;DR enables us to reduce the large dataset into a small set of high-quality data, which can serve as an alternative pre-training dataset. This algorithm significantly speeds up the time-consuming pretraining process. Specifically, TL;DR can compress the mainstream VLP datasets at a high ratio, e.g., reduce well-cleaned CC3M dataset from 2.82M to 0.67M ($\sim$24\%) and noisy YFCC15M from 15M to 2.5M ($\sim$16.7\%). Extensive experiments with three popular VLP models over seven downstream tasks show that VLP model trained on the compressed dataset provided by TL;DR can perform similar or even better results compared with training on the full-scale dataset. The code will be made available at \url{https://github.com/showlab/data-centric.vlp}.
comment: Work in progress. Code: https://github.com/showlab/data-centric.vlp
☆ Understanding and Mitigating Copying in Diffusion Models
Images generated by diffusion models like Stable Diffusion are increasingly widespread. Recent works and even lawsuits have shown that these models are prone to replicating their training data, unbeknownst to the user. In this paper, we first analyze this memorization problem in text-to-image diffusion models. While it is widely believed that duplicated images in the training set are responsible for content replication at inference time, we observe that the text conditioning of the model plays a similarly important role. In fact, we see in our experiments that data replication often does not happen for unconditional models, while it is common in the text-conditional case. Motivated by our findings, we then propose several techniques for reducing data replication at both training and inference time by randomizing and augmenting image captions in the training set.
comment: 17 pages, preprint. Code is available at https://github.com/somepago/DCR
☆ Control4D: Dynamic Portrait Editing by Learning 4D GAN from 2D Diffusion-based Editor
Recent years have witnessed considerable achievements in editing images with text instructions. When applying these editors to dynamic scene editing, the new-style scene tends to be temporally inconsistent due to the frame-by-frame nature of these 2D editors. To tackle this issue, we propose Control4D, a novel approach for high-fidelity and temporally consistent 4D portrait editing. Control4D is built upon an efficient 4D representation with a 2D diffusion-based editor. Instead of using direct supervisions from the editor, our method learns a 4D GAN from it and avoids the inconsistent supervision signals. Specifically, we employ a discriminator to learn the generation distribution based on the edited images and then update the generator with the discrimination signals. For more stable training, multi-level information is extracted from the edited images and used to facilitate the learning of the generator. Experimental results show that Control4D surpasses previous approaches and achieves more photo-realistic and consistent 4D editing performances. The link to our project website is https://control4darxiv.github.io.
comment: The link to our project website is https://control4darxiv.github.io
☆ Feature Learning in Image Hierarchies using Functional Maximal Correlation
This paper proposes the Hierarchical Functional Maximal Correlation Algorithm (HFMCA), a hierarchical methodology that characterizes dependencies across two hierarchical levels in multiview systems. By framing view similarities as dependencies and ensuring contrastivity by imposing orthonormality, HFMCA achieves faster convergence and increased stability in self-supervised learning. HFMCA defines and measures dependencies within image hierarchies, from pixels and patches to full images. We find that the network topology for approximating orthonormal basis functions aligns with a vanilla CNN, enabling the decomposition of density ratios between neighboring layers of feature maps. This approach provides powerful interpretability, revealing the resemblance between supervision and self-supervision through the lens of internal representations.
☆ Chatting Makes Perfect -- Chat-based Image Retrieval
Chats emerge as an effective user-friendly approach for information retrieval, and are successfully employed in many domains, such as customer service, healthcare, and finance. However, existing image retrieval approaches typically address the case of a single query-to-image round, and the use of chats for image retrieval has been mostly overlooked. In this work, we introduce ChatIR: a chat-based image retrieval system that engages in a conversation with the user to elicit information, in addition to an initial query, in order to clarify the user's search intent. Motivated by the capabilities of today's foundation models, we leverage Large Language Models to generate follow-up questions to an initial image description. These questions form a dialog with the user in order to retrieve the desired image from a large corpus. In this study, we explore the capabilities of such a system tested on a large dataset and reveal that engaging in a dialog yields significant gains in image retrieval. We start by building an evaluation pipeline from an existing manually generated dataset and explore different modules and training strategies for ChatIR. Our comparison includes strong baselines derived from related applications trained with Reinforcement Learning. Our system is capable of retrieving the target image from a pool of 50K images with over 78% success rate after 5 dialogue rounds, compared to 75% when questions are asked by humans, and 64% for a single shot text-to-image retrieval. Extensive evaluations reveal the strong capabilities and examine the limitations of CharIR under different settings.
☆ Exploring Regions of Interest: Visualizing Histological Image Classification for Breast Cancer using Deep Learning
Computer aided detection and diagnosis systems based on deep learning have shown promising performance in breast cancer detection. However, there are cases where the obtained results lack justification. In this study, our objective is to highlight the regions of interest used by a convolutional neural network (CNN) for classifying histological images as benign or malignant. We compare these regions with the regions identified by pathologists. To achieve this, we employed the VGG19 architecture and tested three visualization methods: Gradient, LRP Z, and LRP Epsilon. Additionally, we experimented with three pixel selection methods: Bins, K-means, and MeanShift. Based on the results obtained, the Gradient visualization method and the MeanShift selection method yielded satisfactory outcomes for visualizing the images.
☆ Cross-Domain Car Detection Model with Integrated Convolutional Block Attention Mechanism
Car detection, particularly through camera vision, has become a major focus in the field of computer vision and has gained widespread adoption. While current car detection systems are capable of good detection, reliable detection can still be challenging due to factors such as proximity between the car, light intensity, and environmental visibility. To address these issues, we propose a cross-domain car detection model that we apply to car recognition for autonomous driving and other areas. Our model includes several novelties: 1)Building a complete cross-domain target detection framework. 2)Developing an unpaired target domain picture generation module with an integrated convolutional attention mechanism. 3)Adopting Generalized Intersection over Union (GIOU) as the loss function of the target detection framework. 4)Designing an object detection model integrated with two-headed Convolutional Block Attention Module(CBAM). 5)Utilizing an effective data enhancement method. To evaluate the model's effectiveness, we performed a reduced will resolution process on the data in the SSLAD dataset and used it as the benchmark dataset for our task. Experimental results show that the performance of the cross-domain car target detection model improves by 40% over the model without our framework, and our improvements have a significant impact on cross-domain car recognition.
☆ Integrated Decision Gradients: Compute Your Attributions Where the Model Makes Its Decision NeurIPS 2023
Attribution algorithms are frequently employed to explain the decisions of neural network models. Integrated Gradients (IG) is an influential attribution method due to its strong axiomatic foundation. The algorithm is based on integrating the gradients along a path from a reference image to the input image. Unfortunately, it can be observed that gradients computed from regions where the output logit changes minimally along the path provide poor explanations for the model decision, which is called the saturation effect problem. In this paper, we propose an attribution algorithm called integrated decision gradients (IDG). The algorithm focuses on integrating gradients from the region of the path where the model makes its decision, i.e., the portion of the path where the output logit rapidly transitions from zero to its final value. This is practically realized by scaling each gradient by the derivative of the output logit with respect to the path. The algorithm thereby provides a principled solution to the saturation problem. Additionally, we minimize the errors within the Riemann sum approximation of the path integral by utilizing non-uniform subdivisions determined by adaptive sampling. In the evaluation on ImageNet, it is demonstrated that IDG outperforms IG, left-IG, guided IG, and adversarial gradient integration both qualitatively and quantitatively using standard insertion and deletion metrics across three common models.
comment: 18 pages, 8 figures, submitted to NeurIPS 2023, the full code implementation of the paper results is located at: https://github.com/chasewalker26/Integrated-Decision-Gradients
☆ A Unified Conditional Framework for Diffusion-based Image Restoration
Diffusion Probabilistic Models (DPMs) have recently shown remarkable performance in image generation tasks, which are capable of generating highly realistic images. When adopting DPMs for image restoration tasks, the crucial aspect lies in how to integrate the conditional information to guide the DPMs to generate accurate and natural output, which has been largely overlooked in existing works. In this paper, we present a unified conditional framework based on diffusion models for image restoration. We leverage a lightweight UNet to predict initial guidance and the diffusion model to learn the residual of the guidance. By carefully designing the basic module and integration module for the diffusion model block, we integrate the guidance and other auxiliary conditional information into every block of the diffusion model to achieve spatially-adaptive generation conditioning. To handle high-resolution images, we propose a simple yet effective inter-step patch-splitting strategy to produce arbitrary-resolution images without grid artifacts. We evaluate our conditional framework on three challenging tasks: extreme low-light denoising, deblurring, and JPEG restoration, demonstrating its significant improvements in perceptual quality and the generalization to restoration tasks.
☆ FD: On understanding the role of deep feature spaces on face generation evaluation
Perceptual metrics, like the Fr\'echet Inception Distance (FID), are widely used to assess the similarity between synthetically generated and ground truth (real) images. The key idea behind these metrics is to compute errors in a deep feature space that captures perceptually and semantically rich image features. Despite their popularity, the effect that different deep features and their design choices have on a perceptual metric has not been well studied. In this work, we perform a causal analysis linking differences in semantic attributes and distortions between face image distributions to Fr\'echet distances (FD) using several popular deep feature spaces. A key component of our analysis is the creation of synthetic counterfactual faces using deep face generators. Our experiments show that the FD is heavily influenced by its feature space's training dataset and objective function. For example, FD using features extracted from ImageNet-trained models heavily emphasize hats over regions like the eyes and mouth. Moreover, FD using features from a face gender classifier emphasize hair length more than distances in an identity (recognition) feature space. Finally, we evaluate several popular face generation models across feature spaces and find that StyleGAN2 consistently ranks higher than other face generators, except with respect to identity (recognition) features. This suggests the need for considering multiple feature spaces when evaluating generative models and using feature spaces that are tuned to nuances of the domain of interest.
comment: Code and dataset to be released soon
☆ LOWA: Localize Objects in the Wild with Attributes
We present LOWA, a novel method for localizing objects with attributes effectively in the wild. It aims to address the insufficiency of current open-vocabulary object detectors, which are limited by the lack of instance-level attribute classification and rare class names. To train LOWA, we propose a hybrid vision-language training strategy to learn object detection and recognition with class names as well as attribute information. With LOWA, users can not only detect objects with class names, but also able to localize objects by attributes. LOWA is built on top of a two-tower vision-language architecture and consists of a standard vision transformer as the image encoder and a similar transformer as the text encoder. To learn the alignment between visual and text inputs at the instance level, we train LOWA with three training steps: object-level training, attribute-aware learning, and free-text joint training of objects and attributes. This hybrid training strategy first ensures correct object detection, then incorporates instance-level attribute information, and finally balances the object class and attribute sensitivity. We evaluate our model performance of attribute classification and attribute localization on the Open-Vocabulary Attribute Detection (OVAD) benchmark and the Visual Attributes in the Wild (VAW) dataset, and experiments indicate strong zero-shot performance. Ablation studies additionally demonstrate the effectiveness of each training step of our approach.
☆ Tree-Ring Watermarks: Fingerprints for Diffusion Images that are Invisible and Robust
Watermarking the outputs of generative models is a crucial technique for tracing copyright and preventing potential harm from AI-generated content. In this paper, we introduce a novel technique called Tree-Ring Watermarking that robustly fingerprints diffusion model outputs. Unlike existing methods that perform post-hoc modifications to images after sampling, Tree-Ring Watermarking subtly influences the entire sampling process, resulting in a model fingerprint that is invisible to humans. The watermark embeds a pattern into the initial noise vector used for sampling. These patterns are structured in Fourier space so that they are invariant to convolutions, crops, dilations, flips, and rotations. After image generation, the watermark signal is detected by inverting the diffusion process to retrieve the noise vector, which is then checked for the embedded signal. We demonstrate that this technique can be easily applied to arbitrary diffusion models, including text-conditioned Stable Diffusion, as a plug-in with negligible loss in FID. Our watermark is semantically hidden in the image space and is far more robust than watermarking alternatives that are currently deployed. Code is available at github.com/YuxinWenRick/tree-ring-watermark.
comment: 16 pages, 8 figures, code is available at https://github.com/YuxinWenRick/tree-ring-watermark
☆ Physics-Informed Ensemble Representation for Light-Field Image Super-Resolution
Recent learning-based approaches have achieved significant progress in light field (LF) image super-resolution (SR) by exploring convolution-based or transformer-based network structures. However, LF imaging has many intrinsic physical priors that have not been fully exploited. In this paper, we analyze the coordinate transformation of the LF imaging process to reveal the geometric relationship in the LF images. Based on such geometric priors, we introduce a new LF subspace of virtual-slit images (VSI) that provide sub-pixel information complementary to sub-aperture images. To leverage the abundant correlation across the four-dimensional data with manageable complexity, we propose learning ensemble representation of all $C_4^2$ LF subspaces for more effective feature extraction. To super-resolve image structures from undersampled LF data, we propose a geometry-aware decoder, named EPIXformer, which constrains the transformer's operational searching regions with a LF physical prior. Experimental results on both spatial and angular SR tasks demonstrate that the proposed method outperforms other state-of-the-art schemes, especially in handling various disparities.
☆ GANDiffFace: Controllable Generation of Synthetic Datasets for Face Recognition with Realistic Variations
Face recognition systems have significantly advanced in recent years, driven by the availability of large-scale datasets. However, several issues have recently came up, including privacy concerns that have led to the discontinuation of well-established public datasets. Synthetic datasets have emerged as a solution, even though current synthesis methods present other drawbacks such as limited intra-class variations, lack of realism, and unfair representation of demographic groups. This study introduces GANDiffFace, a novel framework for the generation of synthetic datasets for face recognition that combines the power of Generative Adversarial Networks (GANs) and Diffusion models to overcome the limitations of existing synthetic datasets. In GANDiffFace, we first propose the use of GANs to synthesize highly realistic identities and meet target demographic distributions. Subsequently, we fine-tune Diffusion models with the images generated with GANs, synthesizing multiple images of the same identity with a variety of accessories, poses, expressions, and contexts. We generate multiple synthetic datasets by changing GANDiffFace settings, and compare their mated and non-mated score distributions with the distributions provided by popular real-world datasets for face recognition, i.e. VGG2 and IJB-C. Our results show the feasibility of the proposed GANDiffFace, in particular the use of Diffusion models to enhance the (limited) intra-class variations provided by GANs towards the level of real-world datasets.
comment: 8 pages, 6 figures
☆ DeepSolo++: Let Transformer Decoder with Explicit Points Solo for Text Spotting CVPR 2023
End-to-end text spotting aims to integrate scene text detection and recognition into a unified framework. Dealing with the relationship between the two sub-tasks plays a pivotal role in designing effective spotters. Although Transformer-based methods eliminate the heuristic post-processing, they still suffer from the synergy issue between the sub-tasks and low training efficiency. In this paper, we present DeepSolo, a simple DETR-like baseline that lets a single decoder with explicit points solo for text detection and recognition simultaneously and efficiently. Technically, for each text instance, we represent the character sequence as ordered points and model them with learnable explicit point queries. After passing a single decoder, the point queries have encoded requisite text semantics and locations. Furthermore, we show the surprisingly good extensibility of our method, in terms of character class, language type, and task. On the one hand, DeepSolo not only performs well in English scenes but also masters the Chinese transcription with complex font structure and a thousand-level character classes. On the other hand, based on the extensibility of DeepSolo, we launch DeepSolo++ for multilingual text spotting, making a further step to let Transformer decoder with explicit points solo for multilingual text detection, recognition, and script identification all at once. Extensive experiments on public benchmarks demonstrate that our simple approach achieves better training efficiency compared with Transformer-based models and outperforms the previous state-of-the-art. In addition, DeepSolo and DeepSolo++ are also compatible with line annotations, which require much less annotation cost than polygons. The code is available at \url{https://github.com/ViTAE-Transformer/DeepSolo}.
comment: The extension of the CVPR 2023 paper (DeepSolo: Let Transformer Decoder with Explicit Points Solo for Text Spotting)
☆ MicroSegNet: A Deep Learning Approach for Prostate Segmentation on Micro-Ultrasound Images
Micro-ultrasound (micro-US) is a novel 29-MHz ultrasound technique that provides 3-4 times higher resolution than traditional ultrasound, delivering comparable accuracy for diagnosing prostate cancer to MRI but at a lower cost. Accurate prostate segmentation is crucial for prostate volume measurement, cancer diagnosis, prostate biopsy, and treatment planning. This paper proposes a deep learning approach for automated, fast, and accurate prostate segmentation on micro-US images. Prostate segmentation on micro-US is challenging due to artifacts and indistinct borders between the prostate, bladder, and urethra in the midline. We introduce MicroSegNet, a multi-scale annotation-guided Transformer UNet model to address this challenge. During the training process, MicroSegNet focuses more on regions that are hard to segment (challenging regions), where expert and non-expert annotations show discrepancies. We achieve this by proposing an annotation-guided cross entropy loss that assigns larger weight to pixels in hard regions and lower weight to pixels in easy regions. We trained our model using micro-US images from 55 patients, followed by evaluation on 20 patients. Our MicroSegNet model achieved a Dice coefficient of 0.942 and a Hausdorff distance of 2.11 mm, outperforming several state-of-the-art segmentation methods, as well as three human annotators with different experience levels. We will make our code and dataset publicly available to promote transparency and collaboration in research.
☆ Treasure in Distribution: A Domain Randomization based Multi-Source Domain Generalization for 2D Medical Image Segmentation MICCAI 2023
Although recent years have witnessed the great success of convolutional neural networks (CNNs) in medical image segmentation, the domain shift issue caused by the highly variable image quality of medical images hinders the deployment of CNNs in real-world clinical applications. Domain generalization (DG) methods aim to address this issue by training a robust model on the source domain, which has a strong generalization ability. Previously, many DG methods based on feature-space domain randomization have been proposed, which, however, suffer from the limited and unordered search space of feature styles. In this paper, we propose a multi-source DG method called Treasure in Distribution (TriD), which constructs an unprecedented search space to obtain the model with strong robustness by randomly sampling from a uniform distribution. To learn the domain-invariant representations explicitly, we further devise a style-mixing strategy in our TriD, which mixes the feature styles by randomly mixing the augmented and original statistics along the channel wise and can be extended to other DG methods. Extensive experiments on two medical segmentation tasks with different modalities demonstrate that our TriD achieves superior generalization performance on unseen target-domain data. Code is available at https://github.com/Chen-Ziyang/TriD.
comment: 12 pages, 4 figures, 8 tables, early accepted by MICCAI 2023
☆ A Geometric Perspective on Diffusion Models
Recent years have witnessed significant progress in developing efficient training and fast sampling approaches for diffusion models. A recent remarkable advancement is the use of stochastic differential equations (SDEs) to describe data perturbation and generative modeling in a unified mathematical framework. In this paper, we reveal several intriguing geometric structures of diffusion models and contribute a simple yet powerful interpretation to their sampling dynamics. Through carefully inspecting a popular variance-exploding SDE and its marginal-preserving ordinary differential equation (ODE) for sampling, we discover that the data distribution and the noise distribution are smoothly connected with an explicit, quasi-linear sampling trajectory, and another implicit denoising trajectory, which even converges faster in terms of visual quality. We also establish a theoretical relationship between the optimal ODE-based sampling and the classic mean-shift (mode-seeking) algorithm, with which we can characterize the asymptotic behavior of diffusion models and identify the score deviation. These new geometric observations enable us to improve previous sampling algorithms, re-examine latent interpolation, as well as re-explain the working principles of distillation-based fast sampling techniques.
☆ Image Registration of In Vivo Micro-Ultrasound and Ex Vivo Pseudo-Whole Mount Histopathology Images of the Prostate: A Proof-of-Concept Study
Early diagnosis of prostate cancer significantly improves a patient's 5-year survival rate. Biopsy of small prostate cancers is improved with image-guided biopsy. MRI-ultrasound fusion-guided biopsy is sensitive to smaller tumors but is underutilized due to the high cost of MRI and fusion equipment. Micro-ultrasound (micro-US), a novel high-resolution ultrasound technology, provides a cost-effective alternative to MRI while delivering comparable diagnostic accuracy. However, the interpretation of micro-US is challenging due to subtle gray scale changes indicating cancer vs normal tissue. This challenge can be addressed by training urologists with a large dataset of micro-US images containing the ground truth cancer outlines. Such a dataset can be mapped from surgical specimens (histopathology) onto micro-US images via image registration. In this paper, we present a semi-automated pipeline for registering in vivo micro-US images with ex vivo whole-mount histopathology images. Our pipeline begins with the reconstruction of pseudo-whole-mount histopathology images and a 3D micro-US volume. Each pseudo-whole-mount histopathology image is then registered with the corresponding axial micro-US slice using a two-stage approach that estimates an affine transformation followed by a deformable transformation. We evaluated our registration pipeline using micro-US and histopathology images from 18 patients who underwent radical prostatectomy. The results showed a Dice coefficient of 0.94 and a landmark error of 2.7 mm, indicating the accuracy of our registration pipeline. This proof-of-concept study demonstrates the feasibility of accurately aligning micro-US and histopathology images. To promote transparency and collaboration in research, we will make our code and dataset publicly available.
☆ Breast Cancer Detection and Diagnosis: A comparative study of state-of-the-arts deep learning architectures
Breast cancer is a prevalent form of cancer among women, with over 1.5 million women being diagnosed each year. Unfortunately, the survival rates for breast cancer patients in certain third-world countries, like South Africa, are alarmingly low, with only 40% of diagnosed patients surviving beyond five years. The inadequate availability of resources, including qualified pathologists, delayed diagnoses, and ineffective therapy planning, contribute to this low survival rate. To address this pressing issue, medical specialists and researchers have turned to domain-specific AI approaches, specifically deep learning models, to develop end-to-end solutions that can be integrated into computer-aided diagnosis (CAD) systems. By improving the workflow of pathologists, these AI models have the potential to enhance the detection and diagnosis of breast cancer. This research focuses on evaluating the performance of various cutting-edge convolutional neural network (CNN) architectures in comparison to a relatively new model called the Vision Trans-former (ViT). The objective is to determine the superiority of these models in terms of their accuracy and effectiveness. The experimental results reveal that the ViT models outperform the other selected state-of-the-art CNN architectures, achieving an impressive accuracy rate of 95.15%. This study signifies a significant advancement in the field, as it explores the utilization of data augmentation and other relevant preprocessing techniques in conjunction with deep learning models for the detection and diagnosis of breast cancer using datasets of Breast Cancer Histopathological Image Classification.
☆ Speaking the Language of Your Listener: Audience-Aware Adaptation via Plug-and-Play Theory of Mind ACL 2023
Dialogue participants may have varying levels of knowledge about the topic under discussion. In such cases, it is essential for speakers to adapt their utterances by taking their audience into account. Yet, it is an open question how such adaptation can be modelled in computational agents. In this paper, we model a visually grounded referential game between a knowledgeable speaker and a listener with more limited visual and linguistic experience. Inspired by psycholinguistic theories, we endow our speaker with the ability to adapt its referring expressions via a simulation module that monitors the effectiveness of planned utterances from the listener's perspective. We propose an adaptation mechanism building on plug-and-play approaches to controlled language generation, where utterance generation is steered on the fly by the simulator without finetuning the speaker's underlying language model. Our results and analyses show that our approach is effective: the speaker's utterances become closer to the listener's domain of expertise, which leads to higher communicative success.
comment: To appear in Findings of ACL 2023
☆ Joint Adaptive Representations for Image-Language Learning
Image-language learning has made unprecedented progress in visual understanding. These developments have come at high costs, as contemporary vision-language models require large model scales and amounts of data. We here propose a much easier recipe for image-language learning, which produces effective models, outperforming bigger and more expensive ones, often trained on orders of magnitude larger datasets. Our key finding is the joint learning of a compact vision and language representation, which adaptively and iteratively fuses the multi-modal features. This results in a more effective image-language learning, greatly lowering the FLOPs by combining and reducing the number of tokens for both text and images, e.g. a 33\% reduction in FLOPs is achieved, compared to baseline fusion techniques used by popular image-language models, while improving performance. This also allows the model to scale without a large increase in FLOPs or memory. In addition, we propose adaptive pre-training data sampling which improves the data efficiency. The proposed approach achieves competitive performance compared to much larger models, and does so with significantly less data and FLOPs. With only 40M training examples and with 39 GFLOPs our lightweight model outperforms many times larger state-of-the-art models of 2-20x more FLOPs and using bigger datasets some of which with close to 1B training examples.
comment: T4V Workshop
☆ MSKdeX: Musculoskeletal (MSK) decomposition from an X-ray image for fine-grained estimation of lean muscle mass and muscle volume MICCAI 2023
Musculoskeletal diseases such as sarcopenia and osteoporosis are major obstacles to health during aging. Although dual-energy X-ray absorptiometry (DXA) and computed tomography (CT) can be used to evaluate musculoskeletal conditions, frequent monitoring is difficult due to the cost and accessibility (as well as high radiation exposure in the case of CT). We propose a method (named MSKdeX) to estimate fine-grained muscle properties from a plain X-ray image, a low-cost, low-radiation, and highly accessible imaging modality, through musculoskeletal decomposition leveraging fine-grained segmentation in CT. We train a multi-channel quantitative image translation model to decompose an X-ray image into projections of CT of individual muscles to infer the lean muscle mass and muscle volume. We propose the object-wise intensity-sum loss, a simple yet surprisingly effective metric invariant to muscle deformation and projection direction, utilizing information in CT and X-ray images collected from the same patient. While our method is basically an unpaired image-to-image translation, we also exploit the nature of the bone's rigidity, which provides the paired data through 2D-3D rigid registration, adding strong pixel-wise supervision in unpaired training. Through the evaluation using a 539-patient dataset, we showed that the proposed method significantly outperformed conventional methods. The average Pearson correlation coefficient between the predicted and CT-derived ground truth metrics was increased from 0.460 to 0.863. We believe our method opened up a new musculoskeletal diagnosis method and has the potential to be extended to broader applications in multi-channel quantitative image translation tasks. Our source code will be released soon.
comment: MICCAI 2023 provisional acceptance
☆ Neural LerPlane Representations for Fast 4D Reconstruction of Deformable Tissues
Reconstructing deformable tissues from endoscopic stereo videos in robotic surgery is crucial for various clinical applications. However, existing methods relying only on implicit representations are computationally expensive and require dozens of hours, which limits further practical applications. To address this challenge, we introduce LerPlane, a novel method for fast and accurate reconstruction of surgical scenes under a single-viewpoint setting. LerPlane treats surgical procedures as 4D volumes and factorizes them into explicit 2D planes of static and dynamic fields, leading to a compact memory footprint and significantly accelerated optimization. The efficient factorization is accomplished by fusing features obtained through linear interpolation of each plane and enables using lightweight neural networks to model surgical scenes. Besides, LerPlane shares static fields, significantly reducing the workload of dynamic tissue modeling. We also propose a novel sample scheme to boost optimization and improve performance in regions with tool occlusion and large motions. Experiments on DaVinci robotic surgery videos demonstrate that LerPlane accelerates optimization by over 100$\times$ while maintaining high quality across various non-rigid deformations, showing significant promise for future intraoperative surgery applications.
comment: 11 pages, 3 fugure
☆ fpgaHART: A toolflow for throughput-oriented acceleration of 3D CNNs for HAR onto FPGAs
Surveillance systems, autonomous vehicles, human monitoring systems, and video retrieval are just few of the many applications in which 3D Convolutional Neural Networks are exploited. However, their extensive use is restricted by their high computational and memory requirements, especially when integrated into systems with limited resources. This study proposes a toolflow that optimises the mapping of 3D CNN models for Human Action Recognition onto FPGA devices, taking into account FPGA resources and off-chip memory characteristics. The proposed system employs Synchronous Dataflow (SDF) graphs to model the designs and introduces transformations to expand and explore the design space, resulting in high-throughput designs. A variety of 3D CNN models were evaluated using the proposed toolflow on multiple FPGA devices, demonstrating its potential to deliver competitive performance compared to earlier hand-tuned and model-specific designs.
comment: 7 pages, 3 figures, 4 tables. arXiv admin note: substantial text overlap with arXiv:2305.18479
☆ Med-UniC: Unifying Cross-Lingual Medical Vision-Language Pre-Training by Diminishing Bias
The scarcity of data presents a critical obstacle to the efficacy of medical visionlanguage pre-training (VLP). A potential solution lies in the combination of datasets from various language communities. Nevertheless, the main challenge stems from the complexity of integrating diverse syntax and semantics, language-specific medical terminology, and culture-specific implicit knowledge. Therefore, one crucial aspect to consider is the presence of community bias caused by different languages. This paper presents a novel framework named Unifying Cross-Lingual Medical Vision-Language Pre-Training (Med-UniC), designed to integrate multimodal medical data from the two most prevalent languages, English and Spanish. Specifically, we propose Cross-lingual Text Alignment Regularization (CTR) to explicitly unify cross-lingual semantic representations of medical reports originating from diverse language communities. CTR is optimized through latent language disentanglement, rendering our optimization objective to not depend on negative samples, thereby significantly mitigating the bias from determining positive-negative sample pairs within analogous medical reports. Furthermore, it ensures that the cross-lingual representation is not biased toward any specific language community. Med-UniC reaches superior performance across 5 medical image tasks and 10 datasets encompassing over 30 diseases, offering a versatile framework for unifying multi-modal medical data within diverse linguistic communities. The experimental outcomes highlight the presence of community bias in cross-lingual VLP. Reducing this bias enhances the performance not only in vision-language tasks but also in uni-modal visual tasks.
comment: Under review
☆ RaSP: Relation-aware Semantic Prior for Weakly Supervised Incremental Segmentation
Class-incremental semantic image segmentation assumes multiple model updates, each enriching the model to segment new categories. This is typically carried out by providing expensive pixel-level annotations to the training algorithm for all new objects, limiting the adoption of such methods in practical applications. Approaches that solely require image-level labels offer an attractive alternative, yet, such coarse annotations lack precise information about the location and boundary of the new objects. In this paper we argue that, since classes represent not just indices but semantic entities, the conceptual relationships between them can provide valuable information that should be leveraged. We propose a weakly supervised approach that exploits such semantic relations to transfer objectness prior from the previously learned classes into the new ones, complementing the supervisory signal from image-level labels. We validate our approach on a number of continual learning tasks, and show how even a simple pairwise interaction between classes can significantly improve the segmentation mask quality of both old and new classes. We show these conclusions still hold for longer and, hence, more realistic sequences of tasks and for a challenging few-shot scenario.
comment: Accepted to CoLLAs 2023
☆ Unsupervised Anomaly Detection in Medical Images Using Masked Diffusion Model
It can be challenging to identify brain MRI anomalies using supervised deep-learning techniques due to anatomical heterogeneity and the requirement for pixel-level labeling. Unsupervised anomaly detection approaches provide an alternative solution by relying only on sample-level labels of healthy brains to generate a desired representation to identify abnormalities at the pixel level. Although, generative models are crucial for generating such anatomically consistent representations of healthy brains, accurately generating the intricate anatomy of the human brain remains a challenge. In this study, we present a method called masked-DDPM (mDPPM), which introduces masking-based regularization to reframe the generation task of diffusion models. Specifically, we introduce Masked Image Modeling (MIM) and Masked Frequency Modeling (MFM) in our self-supervised approach that enables models to learn visual representations from unlabeled data. To the best of our knowledge, this is the first attempt to apply MFM in DPPM models for medical applications. We evaluate our approach on datasets containing tumors and numerous sclerosis lesions and exhibit the superior performance of our unsupervised method as compared to the existing fully/weakly supervised baselines. Code is available at https://github.com/hasan1292/mDDPM.
Self-supervised Learning to Bring Dual Reversed Rolling Shutter Images Alive SC
Modern consumer cameras usually employ the rolling shutter (RS) mechanism, where images are captured by scanning scenes row-by-row, yielding RS distortions for dynamic scenes. To correct RS distortions, existing methods adopt a fully supervised learning manner, where high framerate global shutter (GS) images should be collected as ground-truth supervision. In this paper, we propose a Self-supervised learning framework for Dual reversed RS distortions Correction (SelfDRSC), where a DRSC network can be learned to generate a high framerate GS video only based on dual RS images with reversed distortions. In particular, a bidirectional distortion warping module is proposed for reconstructing dual reversed RS images, and then a self-supervised loss can be deployed to train DRSC network by enhancing the cycle consistency between input and reconstructed dual reversed RS images. Besides start and end RS scanning time, GS images at arbitrary intermediate scanning time can also be supervised in SelfDRSC, thus enabling the learned DRSC network to generate a high framerate GS video. Moreover, a simple yet effective self-distillation strategy is introduced in self-supervised loss for mitigating boundary artifacts in generated GS images. On synthetic dataset, SelfDRSC achieves better or comparable quantitative metrics in comparison to state-of-the-art methods trained in the full supervision manner. On real-world RS cases, our SelfDRSC can produce high framerate GS videos with finer correction textures and better temporary consistency. The source code and trained models are made publicly available at https://github.com/shangwei5/SelfDRSC.
comment: 16 pages, 16 figures, available at https://github.com/shangwei5/SelfDRSC
☆ Enhancing image quality prediction with self-supervised visual masking
Full-reference image quality metrics (FR-IQMs) aim to measure the visual differences between a pair of reference and distorted images, with the goal of accurately predicting human judgments. However, existing FR-IQMs, including traditional ones like PSNR and SSIM and even perceptual ones such as HDR-VDP, LPIPS, and DISTS, still fall short in capturing the complexities and nuances of human perception. In this work, rather than devising a novel IQM model, we seek to improve upon the perceptual quality of existing FR-IQM methods. We achieve this by considering visual masking, an important characteristic of the human visual system that changes its sensitivity to distortions as a function of local image content. Specifically, for a given FR-IQM metric, we propose to predict a visual masking model that modulates reference and distorted images in a way that penalizes the visual errors based on their visibility. Since the ground truth visual masks are difficult to obtain, we demonstrate how they can be derived in a self-supervised manner solely based on mean opinion scores (MOS) collected from an FR-IQM dataset. Our approach results in enhanced FR-IQM metrics that are more in line with human prediction both visually and quantitatively.
comment: 11 pages, 11 figures
☆ Learning Task-preferred Inference Routes for Gradient De-conflict in Multi-output DNNs
Multi-output deep neural networks(MONs) contain multiple task branches, and these tasks usually share partial network filters that lead to the entanglement of different task inference routes. Due to the inconsistent optimization objectives, the task gradients used for training MONs will interfere with each other on the shared routes, which will decrease the overall model performance. To address this issue, we propose a novel gradient de-conflict algorithm named DR-MGF(Dynamic Routes and Meta-weighted Gradient Fusion) in this work. Different from existing de-conflict methods, DR-MGF achieves gradient de-conflict in MONs by learning task-preferred inference routes. The proposed method is motivated by our experimental findings: the shared filters are not equally important to different tasks. By designing the learnable task-specific importance variables, DR-MGF evaluates the importance of filters for different tasks. Through making the dominances of tasks over filters be proportional to the task-specific importance of filters, DR-MGF can effectively reduce the inter-task interference. The task-specific importance variables ultimately determine task-preferred inference routes at the end of training iterations. Extensive experimental results on CIFAR, ImageNet, and NYUv2 illustrate that DR-MGF outperforms the existing de-conflict methods both in prediction accuracy and convergence speed of MONs. Furthermore, DR-MGF can be extended to general MONs without modifying the overall network structures.
comment: 15 pages
☆ LMCap: Few-shot Multilingual Image Captioning by Retrieval Augmented Language Model Prompting ACL 2023
Multilingual image captioning has recently been tackled by training with large-scale machine translated data, which is an expensive, noisy, and time-consuming process. Without requiring any multilingual caption data, we propose LMCap, an image-blind few-shot multilingual captioning model that works by prompting a language model with retrieved captions. Specifically, instead of following the standard encoder-decoder paradigm, given an image, LMCap first retrieves the captions of similar images using a multilingual CLIP encoder. These captions are then combined into a prompt for an XGLM decoder, in order to generate captions in the desired language. In other words, the generation model does not directly process the image, instead processing retrieved captions. Experiments on the XM3600 dataset of geographically diverse images show that our model is competitive with fully-supervised multilingual captioning models, without requiring any supervised training on any captioning data.
comment: To appear in the Findings of ACL 2023
☆ A Survey of Label-Efficient Deep Learning for 3D Point Clouds
In the past decade, deep neural networks have achieved significant progress in point cloud learning. However, collecting large-scale precisely-annotated training data is extremely laborious and expensive, which hinders the scalability of existing point cloud datasets and poses a bottleneck for efficient exploration of point cloud data in various tasks and applications. Label-efficient learning offers a promising solution by enabling effective deep network training with much-reduced annotation efforts. This paper presents the first comprehensive survey of label-efficient learning of point clouds. We address three critical questions in this emerging research field: i) the importance and urgency of label-efficient learning in point cloud processing, ii) the subfields it encompasses, and iii) the progress achieved in this area. To achieve this, we propose a taxonomy that organizes label-efficient learning methods based on the data prerequisites provided by different types of labels. We categorize four typical label-efficient learning approaches that significantly reduce point cloud annotation efforts: data augmentation, domain transfer learning, weakly-supervised learning, and pretrained foundation models. For each approach, we outline the problem setup and provide an extensive literature review that showcases relevant progress and challenges. Finally, we share insights into current research challenges and potential future directions. A project associated with this survey has been built at \url{https://github.com/xiaoaoran/3D_label_efficient_learning}.
☆ Direct Diffusion Bridge using Data Consistency for Inverse Problems
Diffusion model-based inverse problem solvers have shown impressive performance, but are limited in speed, mostly as they require reverse diffusion sampling starting from noise. Several recent works have tried to alleviate this problem by building a diffusion process, directly bridging the clean and the corrupted for specific inverse problems. In this paper, we first unify these existing works under the name Direct Diffusion Bridges (DDB), showing that while motivated by different theories, the resulting algorithms only differ in the choice of parameters. Then, we highlight a critical limitation of the current DDB framework, namely that it does not ensure data consistency. To address this problem, we propose a modified inference procedure that imposes data consistency without the need for fine-tuning. We term the resulting method data Consistent DDB (CDDB), which outperforms its inconsistent counterpart in terms of both perception and distortion metrics, thereby effectively pushing the Pareto-frontier toward the optimum. Our proposed method achieves state-of-the-art results on both evaluation criteria, showcasing its superiority over existing methods.
comment: 16 pages, 6 figures
☆ Primal-Attention: Self-attention through Asymmetric Kernel SVD in Primal Representation
Recently, a new line of works has emerged to understand and improve self-attention in Transformers by treating it as a kernel machine. However, existing works apply the methods for symmetric kernels to the asymmetric self-attention, resulting in a nontrivial gap between the analytical understanding and numerical implementation. In this paper, we provide a new perspective to represent and optimize self-attention through asymmetric Kernel Singular Value Decomposition (KSVD), which is also motivated by the low-rank property of self-attention normally observed in deep layers. Through asymmetric KSVD, $i$) a primal-dual representation of self-attention is formulated, where the optimization objective is cast to maximize the projection variances in the attention outputs; $ii$) a novel attention mechanism, i.e., Primal-Attention, is proposed via the primal representation of KSVD, avoiding explicit computation of the kernel matrix in the dual; $iii$) with KKT conditions, we prove that the stationary solution to the KSVD optimization in Primal-Attention yields a zero-value objective. In this manner, KSVD optimization can be implemented by simply minimizing a regularization loss, so that low-rank property is promoted without extra decomposition. Numerical experiments show state-of-the-art performance of our Primal-Attention with improved efficiency. Moreover, we demonstrate that the deployed KSVD optimization regularizes Primal-Attention with a sharper singular value decay than that of the canonical self-attention, further verifying the great potential of our method. To the best of our knowledge, this is the first work that provides a primal-dual representation for the asymmetric kernel in self-attention and successfully applies it to modeling and optimization.
comment: We provide a primal-dual representation for the asymmetric self-attention in transformer that allows to avoid explicit computation of the kernel matrix
☆ DeepMerge: Deep Learning-Based Region-Merging for Image Segmentation
Accurate segmentation of large areas from very high spatial-resolution (VHR) remote sensing imagery remains a challenging issue in image analysis. Existing supervised and unsupervised methods both suffer from the large variance of object sizes and the difficulty in scale selection, which often result in poor segmentation accuracies. To address the above challenges, we propose a deep learning-based region-merging method (DeepMerge) to handle the segmentation in large VHR images by integrating a Transformer with a multi-level embedding module, a segment-based feature embedding module and a region-adjacency graph model. In addition, we propose a modified binary tree sampling method to generate multi-level inputs from initial segmentation results, serving as inputs for the DeepMerge model. To our best knowledge, the proposed method is the first to use deep learning to learn the similarity between adjacent segments for region-merging. The proposed DeepMerge method is validated using a remote sensing image of 0.55m resolution covering an area of 5,660 km^2 acquired from Google Earth. The experimental results show that the proposed DeepMerge with the highest F value (0.9446) and the lowest TE (0.0962) and ED2 (0.8989) is able to correctly segment objects of different sizes and outperforms all selected competing segmentation methods from both quantitative and qualitative assessments.
☆ A technique to jointly estimate depth and depth uncertainty for unmanned aerial vehicles
When used by autonomous vehicles for trajectory planning or obstacle avoidance, depth estimation methods need to be reliable. Therefore, estimating the quality of the depth outputs is critical. In this paper, we show how M4Depth, a state-of-the-art depth estimation method designed for unmanned aerial vehicle (UAV) applications, can be enhanced to perform joint depth and uncertainty estimation. For that, we present a solution to convert the uncertainty estimates related to parallax generated by M4Depth into uncertainty estimates related to depth, and show that it outperforms the standard probabilistic approach. Our experiments on various public datasets demonstrate that our method performs consistently, even in zero-shot transfer. Besides, our method offers a compelling value when compared to existing multi-view depth estimation methods as it performs similarly on a multi-view depth estimation benchmark despite being 2.5 times faster and causal, as opposed to other methods. The code of our method is publicly available at https://github.com/michael-fonder/M4DepthU .
comment: The code is available at https://github.com/michael-fonder/M4DepthU
☆ Ambiguity in solving imaging inverse problems with deep learning based operators
In recent years, large convolutional neural networks have been widely used as tools for image deblurring, because of their ability in restoring images very precisely. It is well known that image deblurring is mathematically modeled as an ill-posed inverse problem and its solution is difficult to approximate when noise affects the data. Really, one limitation of neural networks for deblurring is their sensitivity to noise and other perturbations, which can lead to instability and produce poor reconstructions. In addition, networks do not necessarily take into account the numerical formulation of the underlying imaging problem, when trained end-to-end. In this paper, we propose some strategies to improve stability without losing to much accuracy to deblur images with deep-learning based methods. First, we suggest a very small neural architecture, which reduces the execution time for training, satisfying a green AI need, and does not extremely amplify noise in the computed image. Second, we introduce a unified framework where a pre-processing step balances the lack of stability of the following, neural network-based, step. Two different pre-processors are presented: the former implements a strong parameter-free denoiser, and the latter is a variational model-based regularized formulation of the latent imaging problem. This framework is also formally characterized by mathematical analysis. Numerical experiments are performed to verify the accuracy and stability of the proposed approaches for image deblurring when unknown or not-quantified noise is present; the results confirm that they improve the network stability with respect to noise. In particular, the model-based framework represents the most reliable trade-off between visual precision and robustness.
☆ Analytical reconstructions of multiple source-translation computed tomography with extended field of views: a research study
This paper is to investigate the high-quality analytical reconstructions of multiple source-translation computed tomography (mSTCT) under an extended field of view (FOV). Under the larger FOVs, the previously proposed backprojection filtration (BPF) algorithms for mSTCT, including D-BPF and S-BPF, make some intolerable errors in the image edges due to an unstable backprojection weighting factor and the half-scan mode, which deviates from the intention of mSTCT imaging. In this paper, to achieve reconstruction with as little error as possible under the extremely extended FOV, we propose two strategies, including deriving a no-weighting D-BPF (NWD-BPF) for mSTCT and introducing BPFs into a special full-scan mSTCT (F-mSTCT) to balance errors, i.e., abbreviated as FD-BPF and FS-BPF. For the first strategy, we eliminate this unstable backprojection weighting factor by introducing a special variable relationship in D-BPF. For the second strategy, we combine the F-mSTCT geometry with BPFs to study the performance and derive a suitable redundant weighting function for F-mSTCT. The experiments demonstrate our proposed methods for these strategies. Among them, NWD-BPF can weaken the instability at the image edges but blur the details, and FS-BPF can get high-quality stable images under the extremely extended FOV imaging a large object but requires more projections than FD-BPF. For different practical requirements in extending FOV imaging, we give suggestions on algorithm selection.
comment: 17 pages, 11 figures
☆ The Tunnel Effect: Building Data Representations in Deep Neural Networks
Deep neural networks are widely known for their remarkable effectiveness across various tasks, with the consensus that deeper networks implicitly learn more complex data representations. This paper shows that sufficiently deep networks trained for supervised image classification split into two distinct parts that contribute to the resulting data representations differently. The initial layers create linearly-separable representations, while the subsequent layers, which we refer to as \textit{the tunnel}, compress these representations and have a minimal impact on the overall performance. We explore the tunnel's behavior through comprehensive empirical studies, highlighting that it emerges early in the training process. Its depth depends on the relation between the network's capacity and task complexity. Furthermore, we show that the tunnel degrades out-of-distribution generalization and discuss its implications for continual learning.
☆ Towards Monocular Shape from Refraction BMVC
Refraction is a common physical phenomenon and has long been researched in computer vision. Objects imaged through a refractive object appear distorted in the image as a function of the shape of the interface between the media. This hinders many computer vision applications, but can be utilized for obtaining the geometry of the refractive interface. Previous approaches for refractive surface recovery largely relied on various priors or additional information like multiple images of the analyzed surface. In contrast, we claim that a simple energy function based on Snell's law enables the reconstruction of an arbitrary refractive surface geometry using just a single image and known background texture and geometry. In the case of a single point, Snell's law has two degrees of freedom, therefore to estimate a surface depth, we need additional information. We show that solving for an entire surface at once introduces implicit parameter-free spatial regularization and yields convincing results when an intelligent initial guess is provided. We demonstrate our approach through simulations and real-world experiments, where the reconstruction shows encouraging results in the single-frame monocular setting.
comment: 12 pages, 6 figures, The 32nd British Machine Vision Conference (BMVC)
☆ Data Representations' Study of Latent Image Manifolds ICML 2023
Deep neural networks have been demonstrated to achieve phenomenal success in many domains, and yet their inner mechanisms are not well understood. In this paper, we investigate the curvature of image manifolds, i.e., the manifold deviation from being flat in its principal directions. We find that state-of-the-art trained convolutional neural networks for image classification have a characteristic curvature profile along layers: an initial steep increase, followed by a long phase of a plateau, and followed by another increase. In contrast, this behavior does not appear in untrained networks in which the curvature flattens. We also show that the curvature gap between the last two layers has a strong correlation with the generalization capability of the network. Moreover, we find that the intrinsic dimension of latent codes is not necessarily indicative of curvature. Finally, we observe that common regularization methods such as mixup yield flatter representations when compared to other methods. Our experiments show consistent results over a variety of deep learning architectures and multiple data sets. Our code is publicly available at https://github.com/azencot-group/CRLM
comment: Accepted to ICML 2023
☆ Direct Learning-Based Deep Spiking Neural Networks: A Review
The spiking neural network (SNN), as a promising brain-inspired computational model with binary spike information transmission mechanism, rich spatially-temporal dynamics, and event-driven characteristics, has received extensive attention. However, its intricately discontinuous spike mechanism brings difficulty to the optimization of the deep SNN. Since the surrogate gradient method can greatly mitigate the optimization difficulty and shows great potential in directly training deep SNNs, a variety of direct learning-based deep SNN works have been proposed and achieved satisfying progress in recent years. In this paper, we present a comprehensive survey of these direct learning-based deep SNN works, mainly categorized into accuracy improvement methods, efficiency improvement methods, and temporal dynamics utilization methods. In addition, we also divide these categorizations into finer granularities further to better organize and introduce them. Finally, the challenges and trends that may be faced in future research are prospected.
☆ GaitGS: Temporal Feature Learning in Granularity and Span Dimension for Gait Recognition NeurIPS2023
Gait recognition is an emerging biological recognition technology that identifies and verifies individuals based on their walking patterns. However, many current methods are limited in their use of temporal information. In order to fully harness the potential of gait recognition, it is crucial to consider temporal features at various granularities and spans. Hence, in this paper, we propose a novel framework named GaitGS, which aggregates temporal features in the granularity dimension and span dimension simultaneously. Specifically, Multi-Granularity Feature Extractor (MGFE) is proposed to focus on capturing the micro-motion and macro-motion information at the frame level and unit level respectively. Moreover, we present Multi-Span Feature Learning (MSFL) module to generate global and local temporal representations. On three popular gait datasets, extensive experiments demonstrate the state-of-the-art performance of our method. Our method achieves the Rank-1 accuracies of 92.9% (+0.5%), 52.0% (+1.4%), and 97.5% (+0.8%) on CASIA-B, GREW, and OU-MVLP respectively. The source code will be released soon.
comment: Submitted to NeurIPS2023
☆ Spontaneous symmetry breaking in generative diffusion models
Generative diffusion models have recently emerged as a leading approach for generating high-dimensional data. In this paper, we show that the dynamics of these models exhibit a spontaneous symmetry breaking that divides the generative dynamics into two distinct phases: 1) A linear steady-state dynamics around a central fixed-point and 2) an attractor dynamics directed towards the data manifold. These two "phases" are separated by the change in stability of the central fixed-point, with the resulting window of instability being responsible for the diversity of the generated samples. Using both theoretical and empirical evidence, we show that an accurate simulation of the early dynamics does not significantly contribute to the final generation, since early fluctuations are reverted to the central fixed point. To leverage this insight, we propose a Gaussian late initialization scheme, which significantly improves model performance, achieving up to 3x FID improvements on fast samplers, while also increasing sample diversity (e.g., racial composition of generated CelebA images). Our work offers a new way to understand the generative dynamics of diffusion models that has the potential to bring about higher performance and less biased fast-samplers.
☆ VIPriors 3: Visual Inductive Priors for Data-Efficient Deep Learning Challenges
The third edition of the "VIPriors: Visual Inductive Priors for Data-Efficient Deep Learning" workshop featured four data-impaired challenges, focusing on addressing the limitations of data availability in training deep learning models for computer vision tasks. The challenges comprised of four distinct data-impaired tasks, where participants were required to train models from scratch using a reduced number of training samples. The primary objective was to encourage novel approaches that incorporate relevant inductive biases to enhance the data efficiency of deep learning models. To foster creativity and exploration, participants were strictly prohibited from utilizing pre-trained checkpoints and other transfer learning techniques. Significant advancements were made compared to the provided baselines, where winning solutions surpassed the baselines by a considerable margin in all four tasks. These achievements were primarily attributed to the effective utilization of extensive data augmentation policies, model ensembling techniques, and the implementation of data-efficient training methods, including self-supervised representation learning. This report highlights the key aspects of the challenges and their outcomes.
comment: arXiv admin note: text overlap with arXiv:2201.08625
☆ Signal Is Harder To Learn Than Bias: Debiasing with Focal Loss ICLR 2023
Spurious correlations are everywhere. While humans often do not perceive them, neural networks are notorious for learning unwanted associations, also known as biases, instead of the underlying decision rule. As a result, practitioners are often unaware of the biased decision-making of their classifiers. Such a biased model based on spurious correlations might not generalize to unobserved data, leading to unintended, adverse consequences. We propose Signal is Harder (SiH), a variational-autoencoder-based method that simultaneously trains a biased and unbiased classifier using a novel, disentangling reweighting scheme inspired by the focal loss. Using the unbiased classifier, SiH matches or improves upon the performance of state-of-the-art debiasing methods. To improve the interpretability of our technique, we propose a perturbation scheme in the latent space for visualizing the bias that helps practitioners become aware of the sources of spurious correlations.
comment: Presented at the Domain Generalization Workshop (ICLR 2023)
☆ Unveiling Cross Modality Bias in Visual Question Answering: A Causal View with Possible Worlds VQA
To increase the generalization capability of VQA systems, many recent studies have tried to de-bias spurious language or vision associations that shortcut the question or image to the answer. Despite these efforts, the literature fails to address the confounding effect of vision and language simultaneously. As a result, when they reduce bias learned from one modality, they usually increase bias from the other. In this paper, we first model a confounding effect that causes language and vision bias simultaneously, then propose a counterfactual inference to remove the influence of this effect. The model trained in this strategy can concurrently and efficiently reduce vision and language bias. To the best of our knowledge, this is the first work to reduce biases resulting from confounding effects of vision and language in VQA, leveraging causal explain-away relations. We accompany our method with an explain-away strategy, pushing the accuracy of the questions with numerical answers results compared to existing methods that have been an open problem. The proposed method outperforms the state-of-the-art methods in VQA-CP v2 datasets.
comment: 22 pages
☆ Mask, Stitch, and Re-Sample: Enhancing Robustness and Generalizability in Anomaly Detection through Automatic Diffusion Models
The introduction of diffusion models in anomaly detection has paved the way for more effective and accurate image reconstruction in pathologies. However, the current limitations in controlling noise granularity hinder diffusion models' ability to generalize across diverse anomaly types and compromise the restoration of healthy tissues. To overcome these challenges, we propose AutoDDPM, a novel approach that enhances the robustness of diffusion models. AutoDDPM utilizes diffusion models to generate initial likelihood maps of potential anomalies and seamlessly integrates them with the original image. Through joint noised distribution re-sampling, AutoDDPM achieves harmonization and in-painting effects. Our study demonstrates the efficacy of AutoDDPM in replacing anomalous regions while preserving healthy tissues, considerably surpassing diffusion models' limitations. It also contributes valuable insights and analysis on the limitations of current diffusion models, promoting robust and interpretable anomaly detection in medical imaging - an essential aspect of building autonomous clinical decision systems with higher interpretability.
☆ A Unified Framework for U-Net Design and Analysis
U-Nets are a go-to, state-of-the-art neural architecture across numerous tasks for continuous signals on a square such as images and Partial Differential Equations (PDE), however their design and architecture is understudied. In this paper, we provide a framework for designing and analysing general U-Net architectures. We present theoretical results which characterise the role of the encoder and decoder in a U-Net, their high-resolution scaling limits and their conjugacy to ResNets via preconditioning. We propose Multi-ResNets, U-Nets with a simplified, wavelet-based encoder without learnable parameters. Further, we show how to design novel U-Net architectures which encode function constraints, natural bases, or the geometry of the data. In diffusion models, our framework enables us to identify that high-frequency information is dominated by noise exponentially faster, and show how U-Nets with average pooling exploit this. In our experiments, we demonstrate how Multi-ResNets achieve competitive and often superior performance compared to classical U-Nets in image segmentation, PDE surrogate modelling, and generative modelling with diffusion models. Our U-Net framework paves the way to study the theoretical properties of U-Nets and design natural, scalable neural architectures for a multitude of problems beyond the square.
☆ A Multi-Modal Transformer Network for Action Detection
This paper proposes a novel multi-modal transformer network for detecting actions in untrimmed videos. To enrich the action features, our transformer network utilizes a new multi-modal attention mechanism that computes the correlations between different spatial and motion modalities combinations. Exploring such correlations for actions has not been attempted previously. To use the motion and spatial modality more effectively, we suggest an algorithm that corrects the motion distortion caused by camera movement. Such motion distortion, common in untrimmed videos, severely reduces the expressive power of motion features such as optical flow fields. Our proposed algorithm outperforms the state-of-the-art methods on two public benchmarks, THUMOS14 and ActivityNet. We also conducted comparative experiments on our new instructional activity dataset, including a large set of challenging classroom videos captured from elementary schools.
☆ Point-GCC: Universal Self-supervised 3D Scene Pre-training via Geometry-Color Contrast
Geometry and color information provided by the point clouds are both crucial for 3D scene understanding. Two pieces of information characterize the different aspects of point clouds, but existing methods lack an elaborate design for the discrimination and relevance. Hence we explore a 3D self-supervised paradigm that can better utilize the relations of point cloud information. Specifically, we propose a universal 3D scene pre-training framework via Geometry-Color Contrast (Point-GCC), which aligns geometry and color information using a Siamese network. To take care of actual application tasks, we design (i) hierarchical supervision with point-level contrast and reconstruct and object-level contrast based on the novel deep clustering module to close the gap between pre-training and downstream tasks; (ii) architecture-agnostic backbone to adapt for various downstream models. Benefiting from the object-level representation associated with downstream tasks, Point-GCC can directly evaluate model performance and the result demonstrates the effectiveness of our methods. Transfer learning results on a wide range of tasks also show consistent improvements across all datasets. e.g., new state-of-the-art object detection results on SUN RGB-D and S3DIS datasets. Codes will be released at https://github.com/Asterisci/Point-GCC.
☆ XTransCT: Ultra-Fast Volumetric CT Reconstruction using Two Orthogonal X-Ray Projections via a Transformer Network
Computed tomography (CT) scans offer a detailed, three-dimensional representation of patients' internal organs. However, conventional CT reconstruction techniques necessitate acquiring hundreds or thousands of x-ray projections through a complete rotational scan of the body, making navigation or positioning during surgery infeasible. In image-guided radiation therapy, a method that reconstructs ultra-sparse X-ray projections into CT images, we can exploit the substantially reduced radiation dose and minimize equipment burden for localization and navigation. In this study, we introduce a novel Transformer architecture, termed XTransCT, devised to facilitate real-time reconstruction of CT images from two-dimensional X-ray images. We assess our approach regarding image quality and structural reliability using a dataset of fifty patients, supplied by a hospital, as well as the larger public dataset LIDC-IDRI, which encompasses thousands of patients. Additionally, we validated our algorithm's generalizability on the LNDb dataset. Our findings indicate that our algorithm surpasses other methods in image quality, structural precision, and generalizability. Moreover, in comparison to previous 3D convolution-based approaches, we note a substantial speed increase of approximately 300 $\%$, achieving 44 ms per 3D image reconstruction. To guarantee the replicability of our results, we have made our code publicly available.
☆ Intelligible Lip-to-Speech Synthesis with Speech Units
In this paper, we propose a novel Lip-to-Speech synthesis (L2S) framework, for synthesizing intelligible speech from a silent lip movement video. Specifically, to complement the insufficient supervisory signal of the previous L2S model, we propose to use quantized self-supervised speech representations, named speech units, as an additional prediction target for the L2S model. Therefore, the proposed L2S model is trained to generate multiple targets, mel-spectrogram and speech units. As the speech units are discrete while mel-spectrogram is continuous, the proposed multi-target L2S model can be trained with strong content supervision, without using text-labeled data. Moreover, to accurately convert the synthesized mel-spectrogram into a waveform, we introduce a multi-input vocoder that can generate a clear waveform even from blurry and noisy mel-spectrogram by referring to the speech units. Extensive experimental results confirm the effectiveness of the proposed method in L2S.
comment: Interspeech 2023
☆ Boosting Text-to-Image Diffusion Models with Fine-Grained Semantic Rewards
Recent advances in text-to-image diffusion models have achieved remarkable success in generating high-quality, realistic images from given text prompts. However, previous methods fail to perform accurate modality alignment between text concepts and generated images due to the lack of fine-level semantic guidance that successfully diagnoses the modality discrepancy. In this paper, we propose FineRewards to improve the alignment between text and images in text-to-image diffusion models by introducing two new fine-grained semantic rewards: the caption reward and the Semantic Segment Anything (SAM) reward. From the global semantic view, the caption reward generates a corresponding detailed caption that depicts all important contents in the synthetic image via a BLIP-2 model and then calculates the reward score by measuring the similarity between the generated caption and the given prompt. From the local semantic view, the SAM reward segments the generated images into local parts with category labels, and scores the segmented parts by measuring the likelihood of each category appearing in the prompted scene via a large language model, i.e., Vicuna-7B. Additionally, we adopt an assemble reward-ranked learning strategy to enable the integration of multiple reward functions to jointly guide the model training. Adapting results of text-to-image models on the MS-COCO benchmark show that the proposed semantic reward outperforms other baseline reward functions with a considerable margin on both visual quality and semantic similarity with the input prompt. Moreover, by adopting the assemble reward-ranked learning strategy, we further demonstrate that model performance is further improved when adapting under the unifying of the proposed semantic reward with the current image rewards.
☆ Dense and Aligned Captions (DAC) Promote Compositional Reasoning in VL Models
Vision and Language (VL) models offer an effective method for aligning representation spaces of images and text, leading to numerous applications such as cross-modal retrieval, visual question answering, captioning, and more. However, the aligned image-text spaces learned by all the popular VL models are still suffering from the so-called `object bias' - their representations behave as `bags of nouns', mostly ignoring or downsizing the attributes, relations, and states of objects described/appearing in texts/images. Although some great attempts at fixing these `compositional reasoning' issues were proposed in the recent literature, the problem is still far from being solved. In this paper, we uncover two factors limiting the VL models' compositional reasoning performance. These two factors are properties of the paired VL dataset used for finetuning and pre-training the VL model: (i) the caption quality, or in other words `image-alignment', of the texts; and (ii) the `density' of the captions in the sense of mentioning all the details appearing on the image. We propose a fine-tuning approach for automatically treating these factors leveraging a standard VL dataset (CC3M). Applied to CLIP, we demonstrate its significant compositional reasoning performance increase of up to $\sim27\%$ over the base model, up to $\sim20\%$ over the strongest baseline, and by $6.7\%$ on average.
☆ Neural Kernel Surface Reconstruction CVPR 2023
We present a novel method for reconstructing a 3D implicit surface from a large-scale, sparse, and noisy point cloud. Our approach builds upon the recently introduced Neural Kernel Fields (NKF) representation. It enjoys similar generalization capabilities to NKF, while simultaneously addressing its main limitations: (a) We can scale to large scenes through compactly supported kernel functions, which enable the use of memory-efficient sparse linear solvers. (b) We are robust to noise, through a gradient fitting solve. (c) We minimize training requirements, enabling us to learn from any dataset of dense oriented points, and even mix training data consisting of objects and scenes at different scales. Our method is capable of reconstructing millions of points in a few seconds, and handling very large scenes in an out-of-core fashion. We achieve state-of-the-art results on reconstruction benchmarks consisting of single objects, indoor scenes, and outdoor scenes.
comment: CVPR 2023
☆ Exploring Phonetic Context in Lip Movement for Authentic Talking Face Generation
Talking face generation is the task of synthesizing a natural face synchronous to driving audio. Although much progress has been made in terms of visual quality, lip synchronization, and facial motion of the talking face, current works still struggle to overcome issues of crude and asynchronous lip movement, which can result in puppetry-like animation. We identify that the prior works commonly correlate lip movement with audio at the phone level. However, due to co-articulation, where an isolated phone is influenced by the preceding or following phones, the articulation of a phone varies upon the phonetic context. Therefore, modeling lip motion with the phonetic context can generate more spatio-temporally aligned and stable lip movement. In this respect, we investigate the phonetic context in lip motion for authentic talking face generation. We propose a Context-Aware Lip-Sync framework (CALS), which leverages phonetic context to generate more spatio-temporally aligned and stable lip movement. The CALS comprises an Audio-to-Lip module and a Lip-to-Face module. The former explicitly maps each phone to a contextualized lip motion unit, which guides the latter in synthesizing a target identity with context-aware lip motion. In addition, we introduce a discriminative sync critic that enforces accurate lip displacements within the phonetic context through audio-visual sync loss and visual discriminative sync loss. From extensive experiments on LRW, LRS2, and HDTF datasets, we demonstrate that the proposed CALS effectively enhances spatio-temporal alignment, greatly improving upon the state-of-the-art on visual quality, lip-sync quality, and realness. Finally, we show the authenticity of the generated video through a lip readability test and achieve 97.7% of relative word prediction accuracy to real videos.
Spotlight Attention: Robust Object-Centric Learning With a Spatial Locality Prior NeurIPS 2023
The aim of object-centric vision is to construct an explicit representation of the objects in a scene. This representation is obtained via a set of interchangeable modules called \emph{slots} or \emph{object files} that compete for local patches of an image. The competition has a weak inductive bias to preserve spatial continuity; consequently, one slot may claim patches scattered diffusely throughout the image. In contrast, the inductive bias of human vision is strong, to the degree that attention has classically been described with a spotlight metaphor. We incorporate a spatial-locality prior into state-of-the-art object-centric vision models and obtain significant improvements in segmenting objects in both synthetic and real-world datasets. Similar to human visual attention, the combination of image content and spatial constraints yield robust unsupervised object-centric learning, including less sensitivity to model hyperparameters.
comment: 16 pages, 3 figures, under review at NeurIPS 2023
☆ Inferring and Leveraging Parts from Object Shape for Improving Semantic Image Synthesis CVPR 2023
Despite the progress in semantic image synthesis, it remains a challenging problem to generate photo-realistic parts from input semantic map. Integrating part segmentation map can undoubtedly benefit image synthesis, but is bothersome and inconvenient to be provided by users. To improve part synthesis, this paper presents to infer Parts from Object ShapE (iPOSE) and leverage it for improving semantic image synthesis. However, albeit several part segmentation datasets are available, part annotations are still not provided for many object categories in semantic image synthesis. To circumvent it, we resort to few-shot regime to learn a PartNet for predicting the object part map with the guidance of pre-defined support part maps. PartNet can be readily generalized to handle a new object category when a small number (e.g., 3) of support part maps for this category are provided. Furthermore, part semantic modulation is presented to incorporate both inferred part map and semantic map for image synthesis. Experiments show that our iPOSE not only generates objects with rich part details, but also enables to control the image synthesis flexibly. And our iPOSE performs favorably against the state-of-the-art methods in terms of quantitative and qualitative evaluation. Our code will be publicly available at https://github.com/csyxwei/iPOSE.
comment: CVPR 2023. Code will be released at https://github.com/csyxwei/iPOSE
☆ Improving Handwritten OCR with Training Samples Generated by Glyph Conditional Denoising Diffusion Probabilistic Model
Constructing a highly accurate handwritten OCR system requires large amounts of representative training data, which is both time-consuming and expensive to collect. To mitigate the issue, we propose a denoising diffusion probabilistic model (DDPM) to generate training samples. This model conditions on a printed glyph image and creates mappings between printed characters and handwritten images, thus enabling the generation of photo-realistic handwritten samples with diverse styles and unseen text contents. However, the text contents in synthetic images are not always consistent with the glyph conditional images, leading to unreliable labels of synthetic samples. To address this issue, we further propose a progressive data filtering strategy to add those samples with a high confidence of correctness to the training set. Experimental results on IAM benchmark task show that OCR model trained with augmented DDPM-synthesized training samples can achieve about 45% relative word error rate reduction compared with the one trained on real data only.
☆ Automatic Illumination Spectrum Recovery
We develop a deep learning network to estimate the illumination spectrum of hyperspectral images under various lighting conditions. To this end, a dataset, IllumNet, was created. Images were captured using a Specim IQ camera under various illumination conditions, both indoor and outdoor. Outdoor images were captured in sunny, overcast, and shady conditions and at different times of the day. For indoor images, halogen and LED light sources were used, as well as mixed light sources, mainly halogen or LED and fluorescent. The ResNet18 network was employed in this study, but with the 2D kernel changed to a 3D kernel to suit the spectral nature of the data. As well as fitting the actual illumination spectrum well, the predicted illumination spectrum should also be smooth, and this is achieved by the cubic smoothing spline error cost function. Experimental results indicate that the trained model can infer an accurate estimate of the illumination spectrum.
comment: CSIRO Technical report, 19 pages
☆ Label-Retrieval-Augmented Diffusion Models for Learning from Noisy Labels
Learning from noisy labels is an important and long-standing problem in machine learning for real applications. One of the main research lines focuses on learning a label corrector to purify potential noisy labels. However, these methods typically rely on strict assumptions and are limited to certain types of label noise. In this paper, we reformulate the label-noise problem from a generative-model perspective, $\textit{i.e.}$, labels are generated by gradually refining an initial random guess. This new perspective immediately enables existing powerful diffusion models to seamlessly learn the stochastic generative process. Once the generative uncertainty is modeled, we can perform classification inference using maximum likelihood estimation of labels. To mitigate the impact of noisy labels, we propose the $\textbf{L}$abel-$\textbf{R}$etrieval-$\textbf{A}$ugmented (LRA) diffusion model, which leverages neighbor consistency to effectively construct pseudo-clean labels for diffusion training. Our model is flexible and general, allowing easy incorporation of different types of conditional information, $\textit{e.g.}$, use of pre-trained models, to further boost model performance. Extensive experiments are conducted for evaluation. Our model achieves new state-of-the-art (SOTA) results on all the standard real-world benchmark datasets. Remarkably, by incorporating conditional information from the powerful CLIP model, our method can boost the current SOTA accuracy by 10-20 absolute points in many cases.
☆ Towards Accurate and Reliable Change Detection of Remote Sensing Images via Knowledge Review and Online Uncertainty Estimation
Change detection (CD) is an essential task for various real-world applications, such as urban management and disaster assessment. However, previous methods primarily focus on improving the accuracy of CD, while neglecting the reliability of detection results. In this paper, we propose a novel change detection network, called AR-CDNet, which is able to provide accurate change maps and generate pixel-wise uncertainty. Specifically, an online uncertainty estimation branch is constructed to model the pixel-wise uncertainty, which is supervised by the difference between predicted change maps and corresponding ground truth during the training process. Furthermore, we introduce a knowledge review strategy to distill temporal change knowledge from low-level features to high-level ones, thereby enhancing the discriminability of temporal difference features. Finally, we aggregate the uncertainty-aware features extracted from the online uncertainty estimation branch with multi-level temporal difference features to improve the accuracy of CD. Once trained, our AR-CDNet can provide accurate change maps and evaluate pixel-wise uncertainty without ground truth. Experimental results on two benchmark datasets demonstrate the superior performance of AR-CDNet in the CD task. The demo code for our work will be publicly available at \url{https://github.com/guanyuezhen/AR-CDNet}.
☆ A Unified GAN Framework Regarding Manifold Alignment for Remote Sensing Images Generation
Generative Adversarial Networks (GANs) and their variants have achieved remarkable success on natural images. It aims to approximate the distribution of the training datasets. However, their performance degrades when applied to remote sensing (RS) images, and the discriminator often suffers from the overfitting problem. In this paper, we examine the differences between natural and RS images and find that the intrinsic dimensions of RS images are much lower than those of natural images. Besides, the low-dimensional data manifold of RS images may exacerbate the uneven sampling of training datasets and introduce biased information. The discriminator can easily overfit to the biased training distribution, leading to a faulty generation model, even the mode collapse problem. While existing GANs focus on the general distribution of RS datasets, they often neglect the underlying data manifold. In respond, we propose a learnable information-theoretic measure that preserves the intrinsic structures of the original data, and establish a unified GAN framework for manifold alignment in supervised and unsupervised RS image generation.
☆ Perception and Semantic Aware Regularization for Sequential Confidence Calibration
Deep sequence recognition (DSR) models receive increasing attention due to their superior application to various applications. Most DSR models use merely the target sequences as supervision without considering other related sequences, leading to over-confidence in their predictions. The DSR models trained with label smoothing regularize labels by equally and independently smoothing each token, reallocating a small value to other tokens for mitigating overconfidence. However, they do not consider tokens/sequences correlations that may provide more effective information to regularize training and thus lead to sub-optimal performance. In this work, we find tokens/sequences with high perception and semantic correlations with the target ones contain more correlated and effective information and thus facilitate more effective regularization. To this end, we propose a Perception and Semantic aware Sequence Regularization framework, which explore perceptively and semantically correlated tokens/sequences as regularization. Specifically, we introduce a semantic context-free recognition and a language model to acquire similar sequences with high perceptive similarities and semantic correlation, respectively. Moreover, over-confidence degree varies across samples according to their difficulties. Thus, we further design an adaptive calibration intensity module to compute a difficulty score for each samples to obtain finer-grained regularization. Extensive experiments on canonical sequence recognition tasks, including scene text and speech recognition, demonstrate that our method sets novel state-of-the-art results. Code is available at https://github.com/husterpzh/PSSR.
☆ CVSNet: A Computer Implementation for Central Visual System of The Brain
In computer vision, different basic blocks are created around different matrix operations, and models based on different basic blocks have achieved good results. Good results achieved in vision tasks grants them rationality. However, these experimental-based models also make deep learning long criticized for principle and interpretability. Deep learning originated from the concept of neurons in neuroscience, but recent designs detached natural neural networks except for some simple concepts. In this paper, we build an artificial neural network, CVSNet, which can be seen as a computer implementation for central visual system of the brain. Each block in CVSNet represents the same vision information as that in brains. In CVSNet, blocks differs from each other and visual information flows through three independent pathways and five different blocks. Thus CVSNet is completely different from the design of all previous models, in which basic blocks are repeated to build model and information between channels is mixed at the outset. In ablation experiment, we show the information extracted by blocks in CVSNet and compare with previous networks, proving effectiveness and rationality of blocks in CVSNet from experiment side. And in the experiment of object recognition, CVSNet achieves comparable results to ConvNets, Vision Transformers and MLPs.
☆ Noisy-label Learning with Sample Selection based on Noise Rate Estimate
Noisy-labels are challenging for deep learning due to the high capacity of the deep models that can overfit noisy-label training samples. Arguably the most realistic and coincidentally challenging type of label noise is the instance-dependent noise (IDN), where the labelling errors are caused by the ambivalent information present in the images. The most successful label noise learning techniques to address IDN problems usually contain a noisy-label sample selection stage to separate clean and noisy-label samples during training. Such sample selection depends on a criterion, such as loss or gradient, and on a curriculum to define the proportion of training samples to be classified as clean at each training epoch. Even though the estimated noise rate from the training set appears to be a natural signal to be used in the definition of this curriculum, previous approaches generally rely on arbitrary thresholds or pre-defined selection functions to the best of our knowledge. This paper addresses this research gap by proposing a new noisy-label learning graphical model that can easily accommodate state-of-the-art (SOTA) noisy-label learning methods and provide them with a reliable noise rate estimate to be used in a new sample selection curriculum. We show empirically that our model integrated with many SOTA methods can improve their results in many IDN benchmarks, including synthetic and real-world datasets.
☆ Learning by Aligning 2D Skeleton Sequences in Time
This paper presents a novel self-supervised temporal video alignment framework which is useful for several fine-grained human activity understanding applications. In contrast with the state-of-the-art method of CASA, where sequences of 3D skeleton coordinates are taken directly as input, our key idea is to use sequences of 2D skeleton heatmaps as input. Given 2D skeleton heatmaps, we utilize a video transformer which performs self-attention in the spatial and temporal domains for extracting effective spatiotemporal and contextual features. In addition, we introduce simple heatmap augmentation techniques based on 2D skeletons for self-supervised learning. Despite the lack of 3D information, our approach achieves not only higher accuracy but also better robustness against missing and noisy keypoints than CASA. Extensive evaluations on three public datasets, i.e., Penn Action, IKEA ASM, and H2O, demonstrate that our approach outperforms previous methods in different fine-grained human activity understanding tasks, i.e., phase classification, phase progression, video alignment, and fine-grained frame retrieval.
☆ Permutation-Aware Action Segmentation via Unsupervised Frame-to-Segment Alignment
This paper presents a novel transformer-based framework for unsupervised activity segmentation which leverages not only frame-level cues but also segment-level cues. This is in contrast with previous methods which often rely on frame-level information only. Our approach begins with a frame-level prediction module which estimates framewise action classes via a transformer encoder. The frame-level prediction module is trained in an unsupervised manner via temporal optimal transport. To exploit segment-level information, we introduce a segment-level prediction module and a frame-to-segment alignment module. The former includes a transformer decoder for estimating video transcripts, while the latter matches frame-level features with segment-level features, yielding permutation-aware segmentation results. Moreover, inspired by temporal optimal transport, we develop simple-yet-effective pseudo labels for unsupervised training of the above modules. Our experiments on four public datasets, i.e., 50 Salads, YouTube Instructions, Breakfast, and Desktop Assembly show that our approach achieves comparable or better performance than previous methods in unsupervised activity segmentation.
☆ Synthetic CT Generation from MRI using 3D Transformer-based Denoising Diffusion Model
Magnetic resonance imaging (MRI)-based synthetic computed tomography (sCT) simplifies radiation therapy treatment planning by eliminating the need for CT simulation and error-prone image registration, ultimately reducing patient radiation dose and setup uncertainty. We propose an MRI-to-CT transformer-based denoising diffusion probabilistic model (MC-DDPM) to transform MRI into high-quality sCT to facilitate radiation treatment planning. MC-DDPM implements diffusion processes with a shifted-window transformer network to generate sCT from MRI. The proposed model consists of two processes: a forward process which adds Gaussian noise to real CT scans, and a reverse process in which a shifted-window transformer V-net (Swin-Vnet) denoises the noisy CT scans conditioned on the MRI from the same patient to produce noise-free CT scans. With an optimally trained Swin-Vnet, the reverse diffusion process was used to generate sCT scans matching MRI anatomy. We evaluated the proposed method by generating sCT from MRI on a brain dataset and a prostate dataset. Qualitative evaluation was performed using the mean absolute error (MAE) of Hounsfield unit (HU), peak signal to noise ratio (PSNR), multi-scale Structure Similarity index (MS-SSIM) and normalized cross correlation (NCC) indexes between ground truth CTs and sCTs. MC-DDPM generated brain sCTs with state-of-the-art quantitative results with MAE 43.317 HU, PSNR 27.046 dB, SSIM 0.965, and NCC 0.983. For the prostate dataset, MC-DDPM achieved MAE 59.953 HU, PSNR 26.920 dB, SSIM 0.849, and NCC 0.948. In conclusion, we have developed and validated a novel approach for generating CT images from routine MRIs using a transformer-based DDPM. This model effectively captures the complex relationship between CT and MRI images, allowing for robust and high-quality synthetic CT (sCT) images to be generated in minutes.
♻ ☆ Scalable Performance Analysis for Vision-Language Models
Joint vision-language models have shown great performance over a diverse set of tasks. However, little is known about their limitations, as the high dimensional space learned by these models makes it difficult to identify semantic errors. Recent work has addressed this problem by designing highly controlled probing task benchmarks. Our paper introduces a more scalable solution that relies on already annotated benchmarks. Our method consists of extracting a large set of diverse features from a vision-language benchmark and measuring their correlation with the output of the target model. We confirm previous findings that CLIP behaves like a bag of words model and performs better with nouns and verbs; we also uncover novel insights such as CLIP getting confused by concrete words. Our framework is available at https://github.com/MichiganNLP/Scalable-VLM-Probing and can be used with other multimodal models and benchmarks.
comment: Camera-ready version for *SEM 2023
♻ ☆ Dropout Reduces Underfitting ICML 2023
Introduced by Hinton et al. in 2012, dropout has stood the test of time as a regularizer for preventing overfitting in neural networks. In this study, we demonstrate that dropout can also mitigate underfitting when used at the start of training. During the early phase, we find dropout reduces the directional variance of gradients across mini-batches and helps align the mini-batch gradients with the entire dataset's gradient. This helps counteract the stochasticity of SGD and limit the influence of individual batches on model training. Our findings lead us to a solution for improving performance in underfitting models - early dropout: dropout is applied only during the initial phases of training, and turned off afterwards. Models equipped with early dropout achieve lower final training loss compared to their counterparts without dropout. Additionally, we explore a symmetric technique for regularizing overfitting models - late dropout, where dropout is not used in the early iterations and is only activated later in training. Experiments on ImageNet and various vision tasks demonstrate that our methods consistently improve generalization accuracy. Our results encourage more research on understanding regularization in deep learning and our methods can be useful tools for future neural network training, especially in the era of large data. Code is available at https://github.com/facebookresearch/dropout.
comment: ICML 2023
♻ ☆ Accuracy on the Curve: On the Nonlinear Correlation of ML Performance Between Data Subpopulations ICML 2023
Understanding the performance of machine learning (ML) models across diverse data distributions is critically important for reliable applications. Despite recent empirical studies positing a near-perfect linear correlation between in-distribution (ID) and out-of-distribution (OOD) accuracies, we empirically demonstrate that this correlation is more nuanced under subpopulation shifts. Through rigorous experimentation and analysis across a variety of datasets, models, and training epochs, we demonstrate that OOD performance often has a nonlinear correlation with ID performance in subpopulation shifts. Our findings, which contrast previous studies that have posited a linear correlation in model performance during distribution shifts, reveal a "moon shape" correlation (parabolic uptrend curve) between the test performance on the majority subpopulation and the minority subpopulation. This non-trivial nonlinear correlation holds across model architectures, hyperparameters, training durations, and the imbalance between subpopulations. Furthermore, we found that the nonlinearity of this "moon shape" is causally influenced by the degree of spurious correlations in the training data. Our controlled experiments show that stronger spurious correlation in the training data creates more nonlinear performance correlation. We provide complementary experimental and theoretical analyses for this phenomenon, and discuss its implications for ML reliability and fairness. Our work highlights the importance of understanding the nonlinear effects of model improvement on performance in different subpopulations, and has the potential to inform the development of more equitable and responsible machine learning models.
comment: Accepted to the main conference of ICML 2023
♻ ☆ Simple Disentanglement of Style and Content in Visual Representations ICML
Learning visual representations with interpretable features, i.e., disentangled representations, remains a challenging problem. Existing methods demonstrate some success but are hard to apply to large-scale vision datasets like ImageNet. In this work, we propose a simple post-processing framework to disentangle content and style in learned representations from pre-trained vision models. We model the pre-trained features probabilistically as linearly entangled combinations of the latent content and style factors and develop a simple disentanglement algorithm based on the probabilistic model. We show that the method provably disentangles content and style features and verify its efficacy empirically. Our post-processed features yield significant domain generalization performance improvements when the distribution shift occurs due to style changes or style-related spurious correlations.
comment: International Conference on Machine Learning (ICML) 2023
♻ ☆ A Hybrid Quantum-Classical Approach based on the Hadamard Transform for the Convolutional Layer ICML
In this paper, we propose a novel Hadamard Transform (HT)-based neural network layer for hybrid quantum-classical computing. It implements the regular convolutional layers in the Hadamard transform domain. The idea is based on the HT convolution theorem which states that the dyadic convolution between two vectors is equivalent to the element-wise multiplication of their HT representation. Computing the HT is simply the application of a Hadamard gate to each qubit individually, so the HT computations of our proposed layer can be implemented on a quantum computer. Compared to the regular Conv2D layer, the proposed HT-perceptron layer is computationally more efficient. Compared to a CNN with the same number of trainable parameters and 99.26\% test accuracy, our HT network reaches 99.31\% test accuracy with 57.1\% MACs reduced in the MNIST dataset; and in our ImageNet-1K experiments, our HT-based ResNet-50 exceeds the accuracy of the baseline ResNet-50 by 0.59\% center-crop top-1 accuracy using 11.5\% fewer parameters with 12.6\% fewer MACs.
comment: To be presented at International Conference on Machine Learning (ICML), 2023
♻ ☆ Surgical tool classification and localization: results and methods from the MICCAI 2022 SurgToolLoc challenge
The ability to automatically detect and track surgical instruments in endoscopic videos can enable transformational interventions. Assessing surgical performance and efficiency, identifying skilled tool use and choreography, and planning operational and logistical aspects of OR resources are just a few of the applications that could benefit. Unfortunately, obtaining the annotations needed to train machine learning models to identify and localize surgical tools is a difficult task. Annotating bounding boxes frame-by-frame is tedious and time-consuming, yet large amounts of data with a wide variety of surgical tools and surgeries must be captured for robust training. Moreover, ongoing annotator training is needed to stay up to date with surgical instrument innovation. In robotic-assisted surgery, however, potentially informative data like timestamps of instrument installation and removal can be programmatically harvested. The ability to rely on tool installation data alone would significantly reduce the workload to train robust tool-tracking models. With this motivation in mind we invited the surgical data science community to participate in the challenge, SurgToolLoc 2022. The goal was to leverage tool presence data as weak labels for machine learning models trained to detect tools and localize them in video frames with bounding boxes. We present the results of this challenge along with many of the team's efforts. We conclude by discussing these results in the broader context of machine learning and surgical data science. The training data used for this challenge consisting of 24,695 video clips with tool presence labels is also being released publicly and can be accessed at https://console.cloud.google.com/storage/browser/isi-surgtoolloc-2022.
♻ ☆ Invariant Scattering Transform for Medical Imaging
Over the years, the Invariant Scattering Transform (IST) technique has become popular for medical image analysis, including using wavelet transform computation using Convolutional Neural Networks (CNN) to capture patterns' scale and orientation in the input signal. IST aims to be invariant to transformations that are common in medical images, such as translation, rotation, scaling, and deformation, used to improve the performance in medical imaging applications such as segmentation, classification, and registration, which can be integrated into machine learning algorithms for disease detection, diagnosis, and treatment planning. Additionally, combining IST with deep learning approaches has the potential to leverage their strengths and enhance medical image analysis outcomes. This study provides an overview of IST in medical imaging by considering the types of IST, their application, limitations, and potential scopes for future researchers and practitioners.
comment: Accepted for Springer book chapter for a book "Data-driven approaches to Medical Imaging"
♻ ☆ Few Shot Learning for Medical Imaging: A Comparative Analysis of Methodologies and Formal Mathematical Framework
Deep learning becomes an elevated context regarding disposing of many machine learning tasks and has shown a breakthrough upliftment to extract features from unstructured data. Though this flourishing context is developing in the medical image processing sector, scarcity of problem-dependent training data has become a larger issue in the way of easy application of deep learning in the medical sector. To unravel the confined data source, researchers have developed a model that can solve machine learning problems with fewer data called ``Few shot learning". Few hot learning algorithms determine to solve the data limitation problems by extracting the characteristics from a small dataset through classification and segmentation methods. In the medical sector, there is frequently a shortage of available datasets in respect of some confidential diseases. Therefore, Few shot learning gets the limelight in this data scarcity sector. In this chapter, the background and basic overview of a few shots of learning is represented. Henceforth, the classification of few-shot learning is described also. Even the paper shows a comparison of methodological approaches that are applied in medical image analysis over time. The current advancement in the implementation of few-shot learning concerning medical imaging is illustrated. The future scope of this domain in the medical imaging sector is further described.
comment: Accepted for a Springer book chapter for a book title "Data-driven approaches to Medical Imaging"
♻ ☆ SNeRL: Semantic-aware Neural Radiance Fields for Reinforcement Learning ICML 2023
As previous representations for reinforcement learning cannot effectively incorporate a human-intuitive understanding of the 3D environment, they usually suffer from sub-optimal performances. In this paper, we present Semantic-aware Neural Radiance Fields for Reinforcement Learning (SNeRL), which jointly optimizes semantic-aware neural radiance fields (NeRF) with a convolutional encoder to learn 3D-aware neural implicit representation from multi-view images. We introduce 3D semantic and distilled feature fields in parallel to the RGB radiance fields in NeRF to learn semantic and object-centric representation for reinforcement learning. SNeRL outperforms not only previous pixel-based representations but also recent 3D-aware representations both in model-free and model-based reinforcement learning.
comment: ICML 2023. First two authors contributed equally. Order was determined by coin flip
♻ ☆ ZeroFlow: Fast Zero Label Scene Flow via Distillation
Scene flow estimation is the task of describing the 3D motion field between temporally successive point clouds. State-of-the-art methods use strong priors and test-time optimization techniques, but require on the order of tens of seconds for large-scale point clouds, making them unusable as computer vision primitives for real-time applications such as open world object detection. Feed forward methods are considerably faster, running on the order of tens to hundreds of milliseconds for large-scale point clouds, but require expensive human supervision. To address both limitations, we propose Scene Flow via Distillation, a simple distillation framework that uses a label-free optimization method to produce pseudo-labels to supervise a feed forward model. Our instantiation of this framework, ZeroFlow, produces scene flow estimates in real-time on large-scale point clouds at quality competitive with state-of-the-art methods while using zero human labels. Notably, at test-time ZeroFlow is over 1000$\times$ faster than label-free state-of-the-art optimization-based methods on large-scale point clouds and over 1000$\times$ cheaper to train on unlabeled data compared to the cost of human annotation of that data. To facilitate research reuse, we release our code, trained model weights, and high quality pseudo-labels for the Argoverse 2 and Waymo Open datasets.
comment: 9 pages, 4 pages of Supplemental
♻ ☆ Unlocking Slot Attention by Changing Optimal Transport Costs ICML
Slot attention is a powerful method for object-centric modeling in images and videos. However, its set-equivariance limits its ability to handle videos with a dynamic number of objects because it cannot break ties. To overcome this limitation, we first establish a connection between slot attention and optimal transport. Based on this new perspective we propose MESH (Minimize Entropy of Sinkhorn): a cross-attention module that combines the tiebreaking properties of unregularized optimal transport with the speed of regularized optimal transport. We evaluate slot attention using MESH on multiple object-centric learning benchmarks and find significant improvements over slot attention in every setting.
comment: Published at International Conference on Machine Learning (ICML) 2023
♻ ☆ MDAMF: Reconstruction of Cardiac Cine MRI under Free-breathing using Motion-guided Deformable Alignment and Multi-resolution Fusion
Cardiac cine magnetic resonance imaging not only requires higher imaging speed but also needs to address motion artifacts. Especially in the case of free-breathing, more motion artifacts are inevitably introduced. This poses higher demands on the reconstruction performance of the model and its ability to capture temporal information. Previous methods have not effectively utilized the temporal dimension information to compensate for motion artifacts. In order to fully leverage the spatiotemporal information and reduce the impact of motion artifacts, this paper proposes a motion-guided deformable alignment method with second-order bidirectional propagation. Furthermore, aligning adjacent frames may lead to low accuracy or misalignment issues, which are detrimental to subsequent fusion reconstruction. Previous methods have not sufficiently integrated and corrected the aligned feature information. This paper proposes a multi-resolution fusion method to further correct alignment errors or artifacts. Compared to other advanced methods, the proposed approach achieves better image reconstruction quality in terms of peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), and visual effects. The source code will be made available on https://github.com/GtLinyer/MDAMF.
comment: 13 pages, 5 tables, 12 figures
♻ ☆ Concept Decomposition for Visual Exploration and Inspiration
A creative idea is often born from transforming, combining, and modifying ideas from existing visual examples capturing various concepts. However, one cannot simply copy the concept as a whole, and inspiration is achieved by examining certain aspects of the concept. Hence, it is often necessary to separate a concept into different aspects to provide new perspectives. In this paper, we propose a method to decompose a visual concept, represented as a set of images, into different visual aspects encoded in a hierarchical tree structure. We utilize large vision-language models and their rich latent space for concept decomposition and generation. Each node in the tree represents a sub-concept using a learned vector embedding injected into the latent space of a pretrained text-to-image model. We use a set of regularizations to guide the optimization of the embedding vectors encoded in the nodes to follow the hierarchical structure of the tree. Our method allows to explore and discover new concepts derived from the original one. The tree provides the possibility of endless visual sampling at each node, allowing the user to explore the hidden sub-concepts of the object of interest. The learned aspects in each node can be combined within and across trees to create new visual ideas, and can be used in natural language sentences to apply such aspects to new designs.
comment: https://inspirationtree.github.io/inspirationtree/
♻ ☆ Attend-and-Excite: Attention-Based Semantic Guidance for Text-to-Image Diffusion Models SIGGRAPH 2023
Recent text-to-image generative models have demonstrated an unparalleled ability to generate diverse and creative imagery guided by a target text prompt. While revolutionary, current state-of-the-art diffusion models may still fail in generating images that fully convey the semantics in the given text prompt. We analyze the publicly available Stable Diffusion model and assess the existence of catastrophic neglect, where the model fails to generate one or more of the subjects from the input prompt. Moreover, we find that in some cases the model also fails to correctly bind attributes (e.g., colors) to their corresponding subjects. To help mitigate these failure cases, we introduce the concept of Generative Semantic Nursing (GSN), where we seek to intervene in the generative process on the fly during inference time to improve the faithfulness of the generated images. Using an attention-based formulation of GSN, dubbed Attend-and-Excite, we guide the model to refine the cross-attention units to attend to all subject tokens in the text prompt and strengthen - or excite - their activations, encouraging the model to generate all subjects described in the text prompt. We compare our approach to alternative approaches and demonstrate that it conveys the desired concepts more faithfully across a range of text prompts.
comment: Accepted to SIGGRAPH 2023; Project page available at https://yuval-alaluf.github.io/Attend-and-Excite/
♻ ☆ Generalizable Memory-driven Transformer for Multivariate Long Sequence Time-series Forecasting
Multivariate long sequence time-series forecasting (M-LSTF) is a practical but challenging problem. Unlike traditional timer-series forecasting tasks, M-LSTF tasks are more challenging from two aspects: 1) M-LSTF models need to learn time-series patterns both within and between multiple time features; 2) Under the rolling forecasting setting, the similarity between two consecutive training samples increases with the increasing prediction length, which makes models more prone to overfitting. In this paper, we propose a generalizable memory-driven Transformer to target M-LSTF problems. Specifically, we first propose a global-level memory component to drive the forecasting procedure by integrating multiple time-series features. In addition, we adopt a progressive fashion to train our model to increase its generalizability, in which we gradually introduce Bernoulli noises to training samples. Extensive experiments have been performed on five different datasets across multiple fields. Experimental results demonstrate that our approach can be seamlessly plugged into varying Transformer-based models to improve their performances up to roughly 30%. Particularly, this is the first work to specifically focus on the M-LSTF tasks to the best of our knowledge.
comment: Tech report
♻ ☆ Computationally Efficient 3D MRI Reconstruction with Adaptive MLP MICCAI 2023
Compared with 2D MRI, 3D MRI provides superior volumetric spatial resolution and signal-to-noise ratio. However, it is more challenging to reconstruct 3D MRI images. Current methods are mainly based on convolutional neural networks (CNN) with small kernels, which are difficult to scale up to have sufficient fitting power for 3D MRI reconstruction due to the large image size and GPU memory constraint. Furthermore, MRI reconstruction is a deconvolution problem, which demands long-distance information that is difficult to capture by CNNs with small convolution kernels. The multi-layer perceptron (MLP) can model such long-distance information, but it requires a fixed input size. In this paper, we proposed Recon3DMLP, a hybrid of CNN modules with small kernels for low-frequency reconstruction and adaptive MLP (dMLP) modules with large kernels to boost the high-frequency reconstruction, for 3D MRI reconstruction. We further utilized the circular shift operation based on MRI physics such that dMLP accepts arbitrary image size and can extract global information from the entire FOV. We also propose a GPU memory efficient data fidelity module that can reduce $>$50$\%$ memory. We compared Recon3DMLP with other CNN-based models on a high-resolution (HR) 3D MRI dataset. Recon3DMLP improves HR 3D reconstruction and outperforms several existing CNN-based models under similar GPU memory consumption, which demonstrates that Recon3DMLP is a practical solution for HR 3D MRI reconstruction.
comment: MICCAI 2023 early accept
♻ ☆ Towards a Smaller Student: Capacity Dynamic Distillation for Efficient Image Retrieval CVPR2023
Previous Knowledge Distillation based efficient image retrieval methods employs a lightweight network as the student model for fast inference. However, the lightweight student model lacks adequate representation capacity for effective knowledge imitation during the most critical early training period, causing final performance degeneration. To tackle this issue, we propose a Capacity Dynamic Distillation framework, which constructs a student model with editable representation capacity. Specifically, the employed student model is initially a heavy model to fruitfully learn distilled knowledge in the early training epochs, and the student model is gradually compressed during the training. To dynamically adjust the model capacity, our dynamic framework inserts a learnable convolutional layer within each residual block in the student model as the channel importance indicator. The indicator is optimized simultaneously by the image retrieval loss and the compression loss, and a retrieval-guided gradient resetting mechanism is proposed to release the gradient conflict. Extensive experiments show that our method has superior inference speed and accuracy, e.g., on the VeRi-776 dataset, given the ResNet101 as a teacher, our method saves 67.13% model parameters and 65.67% FLOPs (around 24.13% and 21.94% higher than state-of-the-arts) without sacrificing accuracy (around 2.11% mAP higher than state-of-the-arts).
comment: Accepted by CVPR2023
♻ ☆ The Stable Artist: Steering Semantics in Diffusion Latent Space
Large, text-conditioned generative diffusion models have recently gained a lot of attention for their impressive performance in generating high-fidelity images from text alone. However, achieving high-quality results is almost unfeasible in a one-shot fashion. On the contrary, text-guided image generation involves the user making many slight changes to inputs in order to iteratively carve out the envisioned image. However, slight changes to the input prompt often lead to entirely different images being generated, and thus the control of the artist is limited in its granularity. To provide flexibility, we present the Stable Artist, an image editing approach enabling fine-grained control of the image generation process. The main component is semantic guidance (SEGA) which steers the diffusion process along variable numbers of semantic directions. This allows for subtle edits to images, changes in composition and style, as well as optimization of the overall artistic conception. Furthermore, SEGA enables probing of latent spaces to gain insights into the representation of concepts learned by the model, even complex ones such as 'carbon emission'. We demonstrate the Stable Artist on several tasks, showcasing high-quality image editing and composition.
comment: This is a report of preliminary results. A full version of the paper is available at: arXiv:2301.12247
♻ ☆ ILLUME: Rationalizing Vision-Language Models through Human Interactions ICML
Bootstrapping from pre-trained language models has been proven to be an efficient approach for building vision-language models (VLM) for tasks such as image captioning or visual question answering. However, outputs of these models rarely align with user's rationales for specific answers. In order to improve this alignment and reinforce commonsense reasons, we propose a tuning paradigm based on human interactions with machine-generated data. Our ILLUME executes the following loop: Given an image-question-answer prompt, the VLM samples multiple candidate rationales, and a human critic provides feedback via preference selection, used for fine-tuning. This loop increases the training data and gradually carves out the VLM's rationalization capabilities that are aligned with human intent. Our exhaustive experiments demonstrate that ILLUME is competitive with standard supervised finetuning while using significantly fewer training data and only requiring minimal feedback.
comment: Proceedings of the 40th International Conference on Machine Learning (ICML), 2023
♻ ☆ Can We Scale Transformers to Predict Parameters of Diverse ImageNet Models? ICML 2023
Pretraining a neural network on a large dataset is becoming a cornerstone in machine learning that is within the reach of only a few communities with large-resources. We aim at an ambitious goal of democratizing pretraining. Towards that goal, we train and release a single neural network that can predict high quality ImageNet parameters of other neural networks. By using predicted parameters for initialization we are able to boost training of diverse ImageNet models available in PyTorch. When transferred to other datasets, models initialized with predicted parameters also converge faster and reach competitive final performance.
comment: ICML 2023, camera ready (7 tables with extra results added), code and models are at https://github.com/SamsungSAILMontreal/ghn3
♻ ☆ Towards End-to-End Generative Modeling of Long Videos with Memory-Efficient Bidirectional Transformers
Autoregressive transformers have shown remarkable success in video generation. However, the transformers are prohibited from directly learning the long-term dependency in videos due to the quadratic complexity of self-attention, and inherently suffering from slow inference time and error propagation due to the autoregressive process. In this paper, we propose Memory-efficient Bidirectional Transformer (MeBT) for end-to-end learning of long-term dependency in videos and fast inference. Based on recent advances in bidirectional transformers, our method learns to decode the entire spatio-temporal volume of a video in parallel from partially observed patches. The proposed transformer achieves a linear time complexity in both encoding and decoding, by projecting observable context tokens into a fixed number of latent tokens and conditioning them to decode the masked tokens through the cross-attention. Empowered by linear complexity and bidirectional modeling, our method demonstrates significant improvement over the autoregressive Transformers for generating moderately long videos in both quality and speed. Videos and code are available at https://sites.google.com/view/mebt-cvpr2023 .
♻ ☆ X-Paste: Revisiting Scalable Copy-Paste for Instance Segmentation using CLIP and StableDiffusion ICML 2023
Copy-Paste is a simple and effective data augmentation strategy for instance segmentation. By randomly pasting object instances onto new background images, it creates new training data for free and significantly boosts the segmentation performance, especially for rare object categories. Although diverse, high-quality object instances used in Copy-Paste result in more performance gain, previous works utilize object instances either from human-annotated instance segmentation datasets or rendered from 3D object models, and both approaches are too expensive to scale up to obtain good diversity. In this paper, we revisit Copy-Paste at scale with the power of newly emerged zero-shot recognition models (e.g., CLIP) and text2image models (e.g., StableDiffusion). We demonstrate for the first time that using a text2image model to generate images or zero-shot recognition model to filter noisily crawled images for different object categories is a feasible way to make Copy-Paste truly scalable. To make such success happen, we design a data acquisition and processing framework, dubbed ``X-Paste", upon which a systematic study is conducted. On the LVIS dataset, X-Paste provides impressive improvements over the strong baseline CenterNet2 with Swin-L as the backbone. Specifically, it archives +2.6 box AP and +2.1 mask AP gains on all classes and even more significant gains with +6.8 box AP, +6.5 mask AP on long-tail classes. Our code and models are available at https://github.com/yoctta/XPaste.
comment: ICML 2023, code is available at https://github.com/yoctta/XPaste
♻ ☆ Video frame interpolation for high dynamic range sequences captured with dual-exposure sensors
Video frame interpolation (VFI) enables many important applications that might involve the temporal domain, such as slow motion playback, or the spatial domain, such as stop motion sequences. We are focusing on the former task, where one of the key challenges is handling high dynamic range (HDR) scenes in the presence of complex motion. To this end, we explore possible advantages of dual-exposure sensors that readily provide sharp short and blurry long exposures that are spatially registered and whose ends are temporally aligned. This way, motion blur registers temporally continuous information on the scene motion that, combined with the sharp reference, enables more precise motion sampling within a single camera shot. We demonstrate that this facilitates a more complex motion reconstruction in the VFI task, as well as HDR frame reconstruction that so far has been considered only for the originally captured frames, not in-between interpolated frames. We design a neural network trained in these tasks that clearly outperforms existing solutions. We also propose a metric for scene motion complexity that provides important insights into the performance of VFI methods at the test time.
comment: 13 pages, 10 figures
♻ ☆ IB-RAR: Information Bottleneck as Regularizer for Adversarial Robustness
In this paper, we propose a novel method, IB-RAR, which uses Information Bottleneck (IB) to strengthen adversarial robustness for both adversarial training and non-adversarial-trained methods. We first use the IB theory to build regularizers as learning objectives in the loss function. Then, we filter out unnecessary features of intermediate representation according to their mutual information (MI) with labels, as the network trained with IB provides easily distinguishable MI for its features. Experimental results show that our method can be naturally combined with adversarial training and provides consistently better accuracy on new adversarial examples. Our method improves the accuracy by an average of 3.07% against five adversarial attacks for the VGG16 network, trained with three adversarial training benchmarks and the CIFAR-10 dataset. In addition, our method also provides good robustness for undefended methods, such as training with cross-entropy loss only. Finally, in the absence of adversarial training, the VGG16 network trained using our method and the CIFAR-10 dataset reaches an accuracy of 35.86% against PGD examples, while using all layers reaches 25.61% accuracy.
♻ ☆ Image Restoration with Mean-Reverting Stochastic Differential Equations ICML 2023
This paper presents a stochastic differential equation (SDE) approach for general-purpose image restoration. The key construction consists in a mean-reverting SDE that transforms a high-quality image into a degraded counterpart as a mean state with fixed Gaussian noise. Then, by simulating the corresponding reverse-time SDE, we are able to restore the origin of the low-quality image without relying on any task-specific prior knowledge. Crucially, the proposed mean-reverting SDE has a closed-form solution, allowing us to compute the ground truth time-dependent score and learn it with a neural network. Moreover, we propose a maximum likelihood objective to learn an optimal reverse trajectory that stabilizes the training and improves the restoration results. The experiments show that our proposed method achieves highly competitive performance in quantitative comparisons on image deraining, deblurring, and denoising, setting a new state-of-the-art on two deraining datasets. Finally, the general applicability of our approach is further demonstrated via qualitative results on image super-resolution, inpainting, and dehazing. Code is available at https://github.com/Algolzw/image-restoration-sde.
comment: Accepted by ICML 2023; Project page: https://algolzw.github.io/ir-sde/index.html
♻ ☆ Hybrid Energy Based Model in the Feature Space for Out-of-Distribution Detection
Out-of-distribution (OOD) detection is a critical requirement for the deployment of deep neural networks. This paper introduces the HEAT model, a new post-hoc OOD detection method estimating the density of in-distribution (ID) samples using hybrid energy-based models (EBM) in the feature space of a pre-trained backbone. HEAT complements prior density estimators of the ID density, e.g. parametric models like the Gaussian Mixture Model (GMM), to provide an accurate yet robust density estimation. A second contribution is to leverage the EBM framework to provide a unified density estimation and to compose several energy terms. Extensive experiments demonstrate the significance of the two contributions. HEAT sets new state-of-the-art OOD detection results on the CIFAR-10 / CIFAR-100 benchmark as well as on the large-scale Imagenet benchmark. The code is available at: https://github.com/MarcLafon/heatood.
♻ ☆ TypeFormer: Transformers for Mobile Keystroke Biometrics
The broad usage of mobile devices nowadays, the sensitiveness of the information contained in them, and the shortcomings of current mobile user authentication methods are calling for novel, secure, and unobtrusive solutions to verify the users' identity. In this article, we propose TypeFormer, a novel Transformer architecture to model free-text keystroke dynamics performed on mobile devices for the purpose of user authentication. The proposed model consists in Temporal and Channel Modules enclosing two Long Short-Term Memory (LSTM) recurrent layers, Gaussian Range Encoding (GRE), a multi-head Self-Attention mechanism, and a Block-Recurrent structure. Experimenting on one of the largest public databases to date, the Aalto mobile keystroke database, TypeFormer outperforms current state-of-the-art systems achieving Equal Error Rate (EER) values of 3.25% using only 5 enrolment sessions of 50 keystrokes each. In such way, we contribute to reducing the traditional performance gap of the challenging mobile free-text scenario with respect to its desktop and fixed-text counterparts. Additionally, we analyse the behaviour of the model with different experimental configurations such as the length of the keystroke sequences and the amount of enrolment sessions, showing margin for improvement with more enrolment data. Finally, a cross-database evaluation is carried out, demonstrating the robustness of the features extracted by TypeFormer in comparison with existing approaches.
♻ ☆ SPAC-Net: Synthetic Pose-aware Animal ControlNet for Enhanced Pose Estimation
Animal pose estimation has become a crucial area of research, but the scarcity of annotated data is a significant challenge in developing accurate models. Synthetic data has emerged as a promising alternative, but it frequently exhibits domain discrepancies with real data. Style transfer algorithms have been proposed to address this issue, but they suffer from insufficient spatial correspondence, leading to the loss of label information. In this work, we present a new approach called Synthetic Pose-aware Animal ControlNet (SPAC-Net), which incorporates ControlNet into the previously proposed Prior-Aware Synthetic animal data generation (PASyn) pipeline. We leverage the plausible pose data generated by the Variational Auto-Encoder (VAE)-based data generation pipeline as input for the ControlNet Holistically-nested Edge Detection (HED) boundary task model to generate synthetic data with pose labels that are closer to real data, making it possible to train a high-precision pose estimation network without the need for real data. In addition, we propose the Bi-ControlNet structure to separately detect the HED boundary of animals and backgrounds, improving the precision and stability of the generated data. Using the SPAC-Net pipeline, we generate synthetic zebra and rhino images and test them on the AP10K real dataset, demonstrating superior performance compared to using only real images or synthetic data generated by other methods. Our work demonstrates the potential for synthetic data to overcome the challenge of limited annotated data in animal pose estimation.
comment: arXiv admin note: text overlap with arXiv:2208.13944
♻ ☆ Adversarial Detection: Attacking Object Detection in Real Time
Intelligent robots rely on object detection models to perceive the environment. Following advances in deep learning security it has been revealed that object detection models are vulnerable to adversarial attacks. However, prior research primarily focuses on attacking static images or offline videos. Therefore, it is still unclear if such attacks could jeopardize real-world robotic applications in dynamic environments. This paper bridges this gap by presenting the first real-time online attack against object detection models. We devise three attacks that fabricate bounding boxes for nonexistent objects at desired locations. The attacks achieve a success rate of about 90% within about 20 iterations. The demo video is available at https://youtu.be/zJZ1aNlXsMU.
comment: Accepted by IEEE Intelligent Vehicle Symposium, 2023
♻ ☆ Adversarial Driving: Attacking End-to-End Autonomous Driving
As research in deep neural networks advances, deep convolutional networks become promising for autonomous driving tasks. In particular, there is an emerging trend of employing end-to-end neural network models for autonomous driving. However, previous research has shown that deep neural network classifiers are vulnerable to adversarial attacks. While for regression tasks, the effect of adversarial attacks is not as well understood. In this research, we devise two white-box targeted attacks against end-to-end autonomous driving models. Our attacks manipulate the behavior of the autonomous driving system by perturbing the input image. In an average of 800 attacks with the same attack strength (epsilon=1), the image-specific and image-agnostic attack deviates the steering angle from the original output by 0.478 and 0.111, respectively, which is much stronger than random noises that only perturbs the steering angle by 0.002 (The steering angle ranges from [-1, 1]). Both attacks can be initiated in real-time on CPUs without employing GPUs. Demo video: https://youtu.be/I0i8uN2oOP0.
comment: Accepted by IEEE Intelligent Vehicle Symposium, 2023
♻ ☆ The Curse of Recursion: Training on Generated Data Makes Models Forget
Stable Diffusion revolutionised image creation from descriptive text. GPT-2, GPT-3(.5) and GPT-4 demonstrated astonishing performance across a variety of language tasks. ChatGPT introduced such language models to the general public. It is now clear that large language models (LLMs) are here to stay, and will bring about drastic change in the whole ecosystem of online text and images. In this paper we consider what the future might hold. What will happen to GPT-{n} once LLMs contribute much of the language found online? We find that use of model-generated content in training causes irreversible defects in the resulting models, where tails of the original content distribution disappear. We refer to this effect as Model Collapse and show that it can occur in Variational Autoencoders, Gaussian Mixture Models and LLMs. We build theoretical intuition behind the phenomenon and portray its ubiquity amongst all learned generative models. We demonstrate that it has to be taken seriously if we are to sustain the benefits of training from large-scale data scraped from the web. Indeed, the value of data collected about genuine human interactions with systems will be increasingly valuable in the presence of content generated by LLMs in data crawled from the Internet.
♻ ☆ UPop: Unified and Progressive Pruning for Compressing Vision-Language Transformers ICML 2023
Real-world data contains a vast amount of multimodal information, among which vision and language are the two most representative modalities. Moreover, increasingly heavier models, \textit{e}.\textit{g}., Transformers, have attracted the attention of researchers to model compression. However, how to compress multimodal models, especially vison-language Transformers, is still under-explored. This paper proposes the \textbf{U}nified and \textbf{P}r\textbf{o}gressive \textbf{P}runing (\textbf{\emph{UPop}}) as a universal vison-language Transformer compression framework, which incorporates 1) unifiedly searching multimodal subnets in a continuous optimization space from the original model, which enables automatic assignment of pruning ratios among compressible modalities and structures; 2) progressively searching and retraining the subnet, which maintains convergence between the search and retrain to attain higher compression ratios. Experiments on various tasks, datasets, and model architectures demonstrate the effectiveness and versatility of the proposed UPop framework. The code is available at https://github.com/sdc17/UPop.
comment: ICML 2023
♻ ☆ CAP: Correlation-Aware Pruning for Highly-Accurate Sparse Vision Models
Driven by significant improvements in architectural design and training pipelines, computer vision has recently experienced dramatic progress in terms of accuracy on classic benchmarks such as ImageNet. These highly-accurate models are challenging to deploy, as they appear harder to compress using standard techniques such as pruning. We address this issue by introducing the Correlation Aware Pruner (CAP), a new unstructured pruning framework which significantly pushes the compressibility limits for state-of-the-art architectures. Our method is based on two technical advancements: a new theoretically-justified pruner, which can handle complex weight correlations accurately and efficiently during the pruning process itself, and an efficient finetuning procedure for post-compression recovery. We validate our approach via extensive experiments on several modern vision models such as Vision Transformers (ViT), modern CNNs, and ViT-CNN hybrids, showing for the first time that these can be pruned to high sparsity levels (e.g. $\geq 75$%) with low impact on accuracy ($\leq 1$% relative drop). Our approach is also compatible with structured pruning and quantization, and can lead to practical speedups of 1.5 to 2.4x without accuracy loss. To further showcase CAP's accuracy and scalability, we use it to show for the first time that extremely-accurate large vision models, trained via self-supervised techniques, can also be pruned to moderate sparsities, with negligible accuracy loss.
♻ ☆ From Model-Based to Data-Driven Simulation: Challenges and Trends in Autonomous Driving
Simulation is an integral part in the process of developing autonomous vehicles and advantageous for training, validation, and verification of driving functions. Even though simulations come with a series of benefits compared to real-world experiments, various challenges still prevent virtual testing from entirely replacing physical test-drives. Our work provides an overview of these challenges with regard to different aspects and types of simulation and subsumes current trends to overcome them. We cover aspects around perception-, behavior- and content-realism as well as general hurdles in the domain of simulation. Among others, we observe a trend of data-driven, generative approaches and high-fidelity data synthesis to increasingly replace model-based simulation.
♻ ☆ UniFormer: Unifying Convolution and Self-attention for Visual Recognition
It is a challenging task to learn discriminative representation from images and videos, due to large local redundancy and complex global dependency in these visual data. Convolution neural networks (CNNs) and vision transformers (ViTs) have been two dominant frameworks in the past few years. Though CNNs can efficiently decrease local redundancy by convolution within a small neighborhood, the limited receptive field makes it hard to capture global dependency. Alternatively, ViTs can effectively capture long-range dependency via self-attention, while blind similarity comparisons among all the tokens lead to high redundancy. To resolve these problems, we propose a novel Unified transFormer (UniFormer), which can seamlessly integrate the merits of convolution and self-attention in a concise transformer format. Different from the typical transformer blocks, the relation aggregators in our UniFormer block are equipped with local and global token affinity respectively in shallow and deep layers, allowing to tackle both redundancy and dependency for efficient and effective representation learning. Finally, we flexibly stack our UniFormer blocks into a new powerful backbone, and adopt it for various vision tasks from image to video domain, from classification to dense prediction. Without any extra training data, our UniFormer achieves 86.3 top-1 accuracy on ImageNet-1K classification. With only ImageNet-1K pre-training, it can simply achieve state-of-the-art performance in a broad range of downstream tasks, e.g., it obtains 82.9/84.8 top-1 accuracy on Kinetics-400/600, 60.9/71.2 top-1 accuracy on Sth-Sth V1/V2 video classification, 53.8 box AP and 46.4 mask AP on COCO object detection, 50.8 mIoU on ADE20K semantic segmentation, and 77.4 AP on COCO pose estimation. We further build an efficient UniFormer with 2-4x higher throughput. Code is available at https://github.com/Sense-X/UniFormer.
comment: 18 pages, 10 figures, 23 tables. This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible
♻ ☆ ViTMatte: Boosting Image Matting with Pretrained Plain Vision Transformers
Recently, plain vision Transformers (ViTs) have shown impressive performance on various computer vision tasks, thanks to their strong modeling capacity and large-scale pretraining. However, they have not yet conquered the problem of image matting. We hypothesize that image matting could also be boosted by ViTs and present a new efficient and robust ViT-based matting system, named ViTMatte. Our method utilizes (i) a hybrid attention mechanism combined with a convolution neck to help ViTs achieve an excellent performance-computation trade-off in matting tasks. (ii) Additionally, we introduce the detail capture module, which just consists of simple lightweight convolutions to complement the detailed information required by matting. To the best of our knowledge, ViTMatte is the first work to unleash the potential of ViT on image matting with concise adaptation. It inherits many superior properties from ViT to matting, including various pretraining strategies, concise architecture design, and flexible inference strategies. We evaluate ViTMatte on Composition-1k and Distinctions-646, the most commonly used benchmark for image matting, our method achieves state-of-the-art performance and outperforms prior matting works by a large margin.
comment: codes: https://github.com/hustvl/ViTMatte
♻ ☆ Dissecting Self-Supervised Learning Methods for Surgical Computer Vision
The field of surgical computer vision has undergone considerable breakthroughs in recent years with the rising popularity of deep neural network-based methods. However, standard fully-supervised approaches for training such models require vast amounts of annotated data, imposing a prohibitively high cost; especially in the clinical domain. Self-Supervised Learning (SSL) methods, which have begun to gain traction in the general computer vision community, represent a potential solution to these annotation costs, allowing to learn useful representations from only unlabeled data. Still, the effectiveness of SSL methods in more complex and impactful domains, such as medicine and surgery, remains limited and unexplored. In this work, we address this critical need by investigating four state-of-the-art SSL methods (MoCo v2, SimCLR, DINO, SwAV) in the context of surgical computer vision. We present an extensive analysis of the performance of these methods on the Cholec80 dataset for two fundamental and popular tasks in surgical context understanding, phase recognition and tool presence detection. We examine their parameterization, then their behavior with respect to training data quantities in semi-supervised settings. Correct transfer of these methods to surgery, as described and conducted in this work, leads to substantial performance gains over generic uses of SSL - up to 7.4% on phase recognition and 20% on tool presence detection - as well as state-of-the-art semi-supervised phase recognition approaches by up to 14%. Further results obtained on a highly diverse selection of surgical datasets exhibit strong generalization properties. The code is available at https://github.com/CAMMA-public/SelfSupSurg.
♻ ☆ CoDEPS: Online Continual Learning for Depth Estimation and Panoptic Segmentation
Operating a robot in the open world requires a high level of robustness with respect to previously unseen environments. Optimally, the robot is able to adapt by itself to new conditions without human supervision, e.g., automatically adjusting its perception system to changing lighting conditions. In this work, we address the task of continual learning for deep learning-based monocular depth estimation and panoptic segmentation in new environments in an online manner. We introduce CoDEPS to perform continual learning involving multiple real-world domains while mitigating catastrophic forgetting by leveraging experience replay. In particular, we propose a novel domain-mixing strategy to generate pseudo-labels to adapt panoptic segmentation. Furthermore, we explicitly address the limited storage capacity of robotic systems by leveraging sampling strategies for constructing a fixed-size replay buffer based on rare semantic class sampling and image diversity. We perform extensive evaluations of CoDEPS on various real-world datasets demonstrating that it successfully adapts to unseen environments without sacrificing performance on previous domains while achieving state-of-the-art results. The code of our work is publicly available at http://codeps.cs.uni-freiburg.de.
comment: Accepted for "Robotics: Science and Systems (RSS) 2023"
♻ ☆ On the Hidden Mystery of OCR in Large Multimodal Models
Large models have recently played a dominant role in natural language processing and multimodal vision-language learning. It remains less explored about their efficacy in text-related visual tasks. We conducted a comprehensive study of existing publicly available multimodal models, evaluating their performance in text recognition (document text, artistic text, handwritten text, scene text), text-based visual question answering (document text, scene text, and bilingual text), key information extraction (receipts, documents, and nutrition facts) and handwritten mathematical expression recognition. Our findings reveal strengths and weaknesses in these models, which primarily rely on semantic understanding for word recognition and exhibit inferior perception of individual character shapes. They also display indifference towards text length and have limited capabilities in detecting fine-grained features in images. Consequently, these results demonstrate that even the current most powerful large multimodal models cannot match domain-specific methods in traditional text tasks and face greater challenges in more complex tasks. Most importantly, the baseline results showcased in this study could provide a foundational framework for the conception and assessment of innovative strategies targeted at enhancing zero-shot multimodal techniques. Evaluation pipeline will be available at https://github.com/Yuliang-Liu/MultimodalOCR.
♻ ☆ Multi-View Masked World Models for Visual Robotic Manipulation ICML 2023
Visual robotic manipulation research and applications often use multiple cameras, or views, to better perceive the world. How else can we utilize the richness of multi-view data? In this paper, we investigate how to learn good representations with multi-view data and utilize them for visual robotic manipulation. Specifically, we train a multi-view masked autoencoder which reconstructs pixels of randomly masked viewpoints and then learn a world model operating on the representations from the autoencoder. We demonstrate the effectiveness of our method in a range of scenarios, including multi-view control and single-view control with auxiliary cameras for representation learning. We also show that the multi-view masked autoencoder trained with multiple randomized viewpoints enables training a policy with strong viewpoint randomization and transferring the policy to solve real-robot tasks without camera calibration and an adaptation procedure. Video demonstrations are available at: https://sites.google.com/view/mv-mwm.
comment: Accepted to ICML 2023. First two authors contributed equally. Project webpage: https://sites.google.com/view/mv-mwm
♻ ☆ CHiLS: Zero-Shot Image Classification with Hierarchical Label Sets ICML 2023
Open vocabulary models (e.g. CLIP) have shown strong performance on zero-shot classification through their ability generate embeddings for each class based on their (natural language) names. Prior work has focused on improving the accuracy of these models through prompt engineering or by incorporating a small amount of labeled downstream data (via finetuning). However, there has been little focus on improving the richness of the class names themselves, which can pose issues when class labels are coarsely-defined and are uninformative. We propose Classification with Hierarchical Label Sets (or CHiLS), an alternative strategy for zero-shot classification specifically designed for datasets with implicit semantic hierarchies. CHiLS proceeds in three steps: (i) for each class, produce a set of subclasses, using either existing label hierarchies or by querying GPT-3; (ii) perform the standard zero-shot CLIP procedure as though these subclasses were the labels of interest; (iii) map the predicted subclass back to its parent to produce the final prediction. Across numerous datasets with underlying hierarchical structure, CHiLS leads to improved accuracy in situations both with and without ground-truth hierarchical information. CHiLS is simple to implement within existing zero-shot pipelines and requires no additional training cost. Code is available at: https://github.com/acmi-lab/CHILS.
comment: Accepted at ICML 2023
♻ ☆ HiFA: High-fidelity Text-to-3D with Advanced Diffusion Guidance
Automatic text-to-3D synthesis has achieved remarkable advancements through the optimization of 3D models. Existing methods commonly rely on pre-trained text-to-image generative models, such as diffusion models, providing scores for 2D renderings of Neural Radiance Fields (NeRFs) and being utilized for optimizing NeRFs. However, these methods often encounter artifacts and inconsistencies across multiple views due to their limited understanding of 3D geometry. To address these limitations, we propose a reformulation of the optimization loss using the diffusion prior. Furthermore, we introduce a novel training approach that unlocks the potential of the diffusion prior. To improve 3D geometry representation, we apply auxiliary depth supervision for NeRF-rendered images and regularize the density field of NeRFs. Extensive experiments demonstrate the superiority of our method over prior works, resulting in advanced photo-realism and improved multi-view consistency.
comment: Project page: https://hifa-team.github.io/HiFA-site/
♻ ☆ Scaling Up Dataset Distillation to ImageNet-1K with Constant Memory ICLR 2023
Dataset distillation methods aim to compress a large dataset into a small set of synthetic samples, such that when being trained on, competitive performances can be achieved compared to regular training on the entire dataset. Among recently proposed methods, Matching Training Trajectories (MTT) achieves state-of-the-art performance on CIFAR-10/100, while having difficulty scaling to ImageNet-1k dataset due to the large memory requirement when performing unrolled gradient computation through back-propagation. Surprisingly, we show that there exists a procedure to exactly calculate the gradient of the trajectory matching loss with constant GPU memory requirement (irrelevant to the number of unrolled steps). With this finding, the proposed memory-efficient trajectory matching method can easily scale to ImageNet-1K with 6x memory reduction while introducing only around 2% runtime overhead than original MTT. Further, we find that assigning soft labels for synthetic images is crucial for the performance when scaling to larger number of categories (e.g., 1,000) and propose a novel soft label version of trajectory matching that facilities better aligning of model training trajectories on large datasets. The proposed algorithm not only surpasses previous SOTA on ImageNet-1K under extremely low IPCs (Images Per Class), but also for the first time enables us to scale up to 50 IPCs on ImageNet-1K. Our method (TESLA) achieves 27.9% testing accuracy, a remarkable +18.2% margin over prior arts.
comment: ICLR 2023 submission link: https://openreview.net/forum?id=dN70O8pmW8
♻ ☆ Hybrid Representation Learning via Epistemic Graph
In recent years, deep models have achieved remarkable success in many vision tasks. Unfortunately, their performance largely depends on intensive training samples. In contrast, human beings typically perform hybrid learning, e.g., spontaneously integrating structured knowledge for cross-domain recognition or on a much smaller amount of data samples for few-shot learning. Thus it is very attractive to extend hybrid learning for the computer vision tasks by seamlessly integrating structured knowledge with data samples to achieve more effective representation learning. However, such a hybrid learning approach remains a great challenge due to the huge gap between the structured knowledge and the deep features (learned from data samples) on both dimensions and knowledge granularity. In this paper, a novel Epistemic Graph Layer (EGLayer) is developed to enable hybrid learning, such that the information can be exchanged more effectively between the deep features and a structured knowledge graph. Our EGLayer is composed of three major parts: (a) a local graph module to establish a local prototypical graph through the learned deep features, i.e., aligning the deep features with the structured knowledge graph at the same granularity; (b) a query aggregation model to aggregate useful information from the local graphs, and using such representations to compute their similarity with global node embeddings for final prediction; and (c) a novel correlation loss function to constrain the linear consistency between the local and global adjacency matrices.
comment: 13 pages
♻ ☆ Volume Feature Rendering for Fast Neural Radiance Field Reconstruction
Neural radiance fields (NeRFs) are able to synthesize realistic novel views from multi-view images captured from distinct positions and perspectives. In NeRF's rendering pipeline, neural networks are used to represent a scene independently or transform queried learnable feature vector of a point to the expected color or density. With the aid of geometry guides either in occupancy grids or proposal networks, the number of neural network evaluations can be reduced from hundreds to dozens in the standard volume rendering framework. Instead of rendering yielded color after neural network evaluation, we propose to render the queried feature vectors of a ray first and then transform the rendered feature vector to the final pixel color by a neural network. This fundamental change to the standard volume rendering framework requires only one single neural network evaluation to render a pixel, which substantially lowers the high computational complexity of the rendering framework attributed to a large number of neural network evaluations. Consequently, we can use a comparably larger neural network to achieve a better rendering quality while maintaining the same training and rendering time costs. Our model achieves the state-of-the-art rendering quality on both synthetic and real-world datasets while requiring a training time of several minutes.
♻ ☆ A Subabdominal MRI Image Segmentation Algorithm Based on Multi-Scale Feature Pyramid Network and Dual Attention Mechanism
This study aimed to solve the semantic gap and misalignment issue between encoding and decoding because of multiple convolutional and pooling operations in U-Net when segmenting subabdominal MRI images during rectal cancer treatment. A MRI Image Segmentation is proposed based on a multi-scale feature pyramid network and dual attention mechanism. Our innovation is the design of two modules: 1) a dilated convolution and multi-scale feature pyramid network are used in the encoding to avoid the semantic gap. 2) a dual attention mechanism is designed to maintain spatial information of U-Net and reduce misalignment. Experiments on a subabdominal MRI image dataset show the proposed method achieves better performance than others methods. In conclusion, a multi-scale feature pyramid network can reduce the semantic gap, and the dual attention mechanism can make an alignment of features between encoding and decoding.
comment: 20 pages,9 figures
♻ ☆ Fine-grained Image Editing by Pixel-wise Guidance Using Diffusion Models CVPR 2023
Our goal is to develop fine-grained real-image editing methods suitable for real-world applications. In this paper, we first summarize four requirements for these methods and propose a novel diffusion-based image editing framework with pixel-wise guidance that satisfies these requirements. Specifically, we train pixel-classifiers with a few annotated data and then infer the segmentation map of a target image. Users then manipulate the map to instruct how the image will be edited. We utilize a pre-trained diffusion model to generate edited images aligned with the user's intention with pixel-wise guidance. The effective combination of proposed guidance and other techniques enables highly controllable editing with preserving the outside of the edited area, which results in meeting our requirements. The experimental results demonstrate that our proposal outperforms the GAN-based method for editing quality and speed.
comment: Accepted by AI for Content Creation (AI4CC) workshop at CVPR 2023
♻ ☆ A Framework For Refining Text Classification and Object Recognition from Academic Articles
With the widespread use of the internet, it has become increasingly crucial to extract specific information from vast amounts of academic articles efficiently. Data mining techniques are generally employed to solve this issue. However, data mining for academic articles is challenging since it requires automatically extracting specific patterns in complex and unstructured layout documents. Current data mining methods for academic articles employ rule-based(RB) or machine learning(ML) approaches. However, using rule-based methods incurs a high coding cost for complex typesetting articles. On the other hand, simply using machine learning methods requires annotation work for complex content types within the paper, which can be costly. Furthermore, only using machine learning can lead to cases where patterns easily recognized by rule-based methods are mistakenly extracted. To overcome these issues, from the perspective of analyzing the standard layout and typesetting used in the specified publication, we emphasize implementing specific methods for specific characteristics in academic articles. We have developed a novel Text Block Refinement Framework (TBRF), a machine learning and rule-based scheme hybrid. We used the well-known ACL proceeding articles as experimental data for the validation experiment. The experiment shows that our approach achieved over 95% classification accuracy and 90% detection accuracy for tables and figures.
comment: This paper will be submitted to 'The International Symposium on Innovations in Intelligent SysTems and Applications 2023 (INISTA 2023)'
♻ ☆ atTRACTive: Semi-automatic white matter tract segmentation using active learning
Accurately identifying white matter tracts in medical images is essential for various applications, including surgery planning and tract-specific analysis. Supervised machine learning models have reached state-of-the-art solving this task automatically. However, these models are primarily trained on healthy subjects and struggle with strong anatomical aberrations, e.g. caused by brain tumors. This limitation makes them unsuitable for tasks such as preoperative planning, wherefore time-consuming and challenging manual delineation of the target tract is typically employed. We propose semi-automatic entropy-based active learning for quick and intuitive segmentation of white matter tracts from whole-brain tractography consisting of millions of streamlines. The method is evaluated on 21 openly available healthy subjects from the Human Connectome Project and an internal dataset of ten neurosurgical cases. With only a few annotations, the proposed approach enables segmenting tracts on tumor cases comparable to healthy subjects (dice=0.71), while the performance of automatic methods, like TractSeg dropped substantially (dice=0.34) in comparison to healthy subjects. The method is implemented as a prototype named atTRACTive in the freely available software MITK Diffusion. Manual experiments on tumor data showed higher efficiency due to lower segmentation times compared to traditional ROI-based segmentation.
♻ ☆ Multi-task Paired Masking with Alignment Modeling for Medical Vision-Language Pre-training
In recent years, the growing demand for medical imaging diagnosis has placed a significant burden on radiologists. As a solution, Medical Vision-Language Pre-training (Med-VLP) methods have been proposed to learn universal representations from medical images and reports, benefiting downstream tasks without requiring fine-grained annotations. However, existing methods have overlooked the importance of cross-modal alignment in joint image-text reconstruction, resulting in insufficient cross-modal interaction. To address this limitation, we propose a unified Med-VLP framework based on Multi-task Paired Masking with Alignment (MPMA) to integrate the cross-modal alignment task into the joint image-text reconstruction framework to achieve more comprehensive cross-modal interaction, while a Global and Local Alignment (GLA) module is designed to assist self-supervised paradigm in obtaining semantic representations with rich domain knowledge. Furthermore, we introduce a Memory-Augmented Cross-Modal Fusion (MA-CMF) module to fully integrate visual information to assist report reconstruction and fuse the multi-modal representations adequately. Experimental results demonstrate that the proposed unified approach outperforms previous methods in all downstream tasks, including uni-modal, cross-modal, and multi-modal tasks.
♻ ☆ Consistency Models ICML 2023
Diffusion models have significantly advanced the fields of image, audio, and video generation, but they depend on an iterative sampling process that causes slow generation. To overcome this limitation, we propose consistency models, a new family of models that generate high quality samples by directly mapping noise to data. They support fast one-step generation by design, while still allowing multistep sampling to trade compute for sample quality. They also support zero-shot data editing, such as image inpainting, colorization, and super-resolution, without requiring explicit training on these tasks. Consistency models can be trained either by distilling pre-trained diffusion models, or as standalone generative models altogether. Through extensive experiments, we demonstrate that they outperform existing distillation techniques for diffusion models in one- and few-step sampling, achieving the new state-of-the-art FID of 3.55 on CIFAR-10 and 6.20 on ImageNet 64x64 for one-step generation. When trained in isolation, consistency models become a new family of generative models that can outperform existing one-step, non-adversarial generative models on standard benchmarks such as CIFAR-10, ImageNet 64x64 and LSUN 256x256.
comment: ICML 2023
♻ ☆ HICO-DET-SG and V-COCO-SG: New Data Splits for Evaluating the Systematic Generalization Performance of Human-Object Interaction Detection Models
Human-Object Interaction (HOI) detection is a task to localize humans and objects in an image and predict the interactions in human-object pairs. In real-world scenarios, HOI detection models are required systematic generalization, i.e., generalization to novel combinations of objects and interactions, because the train data are expected to cover a limited portion of all possible combinations. However, to our knowledge, no open benchmarks or previous work exist for evaluating the systematic generalization performance of HOI detection models. To address this issue, we created two new sets of HOI detection data splits named HICO-DET-SG and V-COCO-SG based on the HICO-DET and V-COCO datasets, respectively. When evaluated on the new data splits, the representative HOI detection models performed much more poorly than when evaluated on the original splits. This reveals that systematic generalization is a challenging goal in HOI detection. By analyzing the evaluation results, we also gain insights for improving the systematic generalization performance and identify four possible future research directions. We hope that our new data splits and presented analysis will encourage further research on systematic generalization in HOI detection.
comment: 19 pages, 3 figures, 4 tables, This is a non-anonymized version of the manuscript submitted to Transactions on Machine Learning Research (TMLR)
♻ ☆ Ultrafast Image Categorization in Biology and Neural Models
Humans are able to categorize images very efficiently, in particular to detect the presence of an animal very quickly. Recently, deep learning algorithms based on convolutional neural networks (CNNs) have achieved higher than human accuracy for a wide range of visual categorization tasks. However, the tasks on which these artificial networks are typically trained and evaluated tend to be highly specialized and do not generalize well, e.g., accuracy drops after image rotation. In this respect, biological visual systems are more flexible and efficient than artificial systems for more general tasks, such as recognizing an animal. To further the comparison between biological and artificial neural networks, we re-trained the standard VGG 16 CNN on two independent tasks that are ecologically relevant to humans: detecting the presence of an animal or an artifact. We show that re-training the network achieves a human-like level of performance, comparable to that reported in psychophysical tasks. In addition, we show that the categorization is better when the outputs of the models are combined. Indeed, animals (e.g., lions) tend to be less present in photographs that contain artifacts (e.g., buildings). Furthermore, these re-trained models were able to reproduce some unexpected behavioral observations from human psychophysics, such as robustness to rotation (e.g., an upside-down or tilted image) or to a grayscale transformation. Finally, we quantified the number of CNN layers required to achieve such performance and showed that good accuracy for ultrafast image categorization can be achieved with only a few layers, challenging the belief that image recognition requires deep sequential analysis of visual objects.
♻ ☆ ImageBind: One Embedding Space To Bind Them All CVPR 2023
We present ImageBind, an approach to learn a joint embedding across six different modalities - images, text, audio, depth, thermal, and IMU data. We show that all combinations of paired data are not necessary to train such a joint embedding, and only image-paired data is sufficient to bind the modalities together. ImageBind can leverage recent large scale vision-language models, and extends their zero-shot capabilities to new modalities just by using their natural pairing with images. It enables novel emergent applications 'out-of-the-box' including cross-modal retrieval, composing modalities with arithmetic, cross-modal detection and generation. The emergent capabilities improve with the strength of the image encoder and we set a new state-of-the-art on emergent zero-shot recognition tasks across modalities, outperforming specialist supervised models. Finally, we show strong few-shot recognition results outperforming prior work, and that ImageBind serves as a new way to evaluate vision models for visual and non-visual tasks.
comment: CVPR 2023 (Highlighted Paper). Website: https://imagebind.metademolab.com/ Code/Models: https://github.com/facebookresearch/ImageBind
♻ ☆ OmniMAE: Single Model Masked Pretraining on Images and Videos CVPR 2023
Transformer-based architectures have become competitive across a variety of visual domains, most notably images and videos. While prior work studies these modalities in isolation, having a common architecture suggests that one can train a single unified model for multiple visual modalities. Prior attempts at unified modeling typically use architectures tailored for vision tasks, or obtain worse performance compared to single modality models. In this work, we show that masked autoencoding can be used to train a simple Vision Transformer on images and videos, without requiring any labeled data. This single model learns visual representations that are comparable to or better than single-modality representations on both image and video benchmarks, while using a much simpler architecture. Furthermore, this model can be learned by dropping 90% of the image and 95% of the video patches, enabling extremely fast training of huge model architectures. In particular, we show that our single ViT-Huge model can be finetuned to achieve 86.6% on ImageNet and 75.5% on the challenging Something Something-v2 video benchmark, setting a new state-of-the-art.
comment: CVPR 2023. Code/models: https://github.com/facebookresearch/omnivore
♻ ☆ MaskedKD: Efficient Distillation of Vision Transformers with Masked Images
Knowledge distillation is an effective method for training lightweight models, but it introduces a significant amount of computational overhead to the training cost, as the method requires acquiring teacher supervisions on training samples. This additional cost -- called distillation cost -- is most pronounced when we employ large-scale teacher models such as vision transformers (ViTs). We present MaskedKD, a simple yet effective strategy that can significantly reduce the cost of distilling ViTs without sacrificing the prediction accuracy of the student model. Specifically, MaskedKD diminishes the cost of running teacher at inference by masking a fraction of image patch tokens fed to the teacher, and therefore skipping the computations required to process those patches. The mask locations are selected to prevent masking away the core features of an image that the student model uses for prediction. This mask selection mechanism operates based on some attention score of the student model, which is already computed during the student forward pass, and thus incurs almost no additional computation. Without sacrificing the final student accuracy, MaskedKD dramatically reduces the amount of computations required for distilling ViTs. We demonstrate that MaskedKD can save up the distillation cost by $50\%$ without any student performance drop, leading to approximately $28\%$ drop in the overall training FLOPs.
♻ ☆ Decomposed Human Motion Prior for Video Pose Estimation via Adversarial Training
Estimating human pose from video is a task that receives considerable attention due to its applicability in numerous 3D fields. The complexity of prior knowledge of human body movements poses a challenge to neural network models in the task of regressing keypoints. In this paper, we address this problem by incorporating motion prior in an adversarial way. Different from previous methods, we propose to decompose holistic motion prior to joint motion prior, making it easier for neural networks to learn from prior knowledge thereby boosting the performance on the task. We also utilize a novel regularization loss to balance accuracy and smoothness introduced by motion prior. Our method achieves 9\% lower PA-MPJPE and 29\% lower acceleration error than previous methods tested on 3DPW. The estimator proves its robustness by achieving impressive performance on in-the-wild dataset.
♻ ☆ Floorplan Restoration by Structure Hallucinating Transformer Cascades
This paper presents an extreme floorplan reconstruction task, a new benchmark for the task, and a neural architecture as a solution. Given a partial floorplan reconstruction inferred or curated from panorama images, the task is to reconstruct a complete floorplan including invisible architectural structures. The proposed neural network 1) encodes an input partial floorplan into a set of latent vectors by convolutional neural networks and a Transformer; and 2) reconstructs an entire floorplan while hallucinating invisible rooms and doors by cascading Transformer decoders. Qualitative and quantitative evaluations demonstrate effectiveness of our approach over the benchmark of 701 houses, outperforming the state-of-the-art reconstruction techniques. We will share our code, models, and data.
♻ ☆ PuzzleFusion: Unleashing the Power of Diffusion Models for Spatial Puzzle Solving
This paper presents an end-to-end neural architecture based on Diffusion Models for spatial puzzle solving, particularly jigsaw puzzle and room arrangement tasks. In the latter task, for instance, the proposed system ``PuzzleFusion'' takes a set of room layouts as polygonal curves in the top-down view and aligns the room layout pieces by estimating their 2D translations and rotations, akin to solving the jigsaw puzzle of room layouts. A surprising discovery of the paper is that the simple use of a Diffusion Model effectively solves these challenging spatial puzzle tasks as a conditional generation process. To enable learning of an end-to-end neural system, the paper introduces new datasets with ground-truth arrangements: 1) 2D Voronoi jigsaw dataset, a synthetic one where pieces are generated by Voronoi diagram of 2D pointset; and 2) MagicPlan dataset, a real one offered by MagicPlan from its production pipeline, where pieces are room layouts constructed by augmented reality App by real-estate consumers. The qualitative and quantitative evaluations demonstrate that our approach outperforms the competing methods by significant margins in all the tasks. We will publicly share all our code and data.
♻ ☆ High-fidelity Generalized Emotional Talking Face Generation with Multi-modal Emotion Space Learning
Recently, emotional talking face generation has received considerable attention. However, existing methods only adopt one-hot coding, image, or audio as emotion conditions, thus lacking flexible control in practical applications and failing to handle unseen emotion styles due to limited semantics. They either ignore the one-shot setting or the quality of generated faces. In this paper, we propose a more flexible and generalized framework. Specifically, we supplement the emotion style in text prompts and use an Aligned Multi-modal Emotion encoder to embed the text, image, and audio emotion modality into a unified space, which inherits rich semantic prior from CLIP. Consequently, effective multi-modal emotion space learning helps our method support arbitrary emotion modality during testing and could generalize to unseen emotion styles. Besides, an Emotion-aware Audio-to-3DMM Convertor is proposed to connect the emotion condition and the audio sequence to structural representation. A followed style-based High-fidelity Emotional Face generator is designed to generate arbitrary high-resolution realistic identities. Our texture generator hierarchically learns flow fields and animated faces in a residual manner. Extensive experiments demonstrate the flexibility and generalization of our method in emotion control and the effectiveness of high-quality face synthesis.
♻ ☆ Statistically Significant Concept-based Explanation of Image Classifiers via Model Knockoffs IJCAI'23
A concept-based classifier can explain the decision process of a deep learning model by human-understandable concepts in image classification problems. However, sometimes concept-based explanations may cause false positives, which misregards unrelated concepts as important for the prediction task. Our goal is to find the statistically significant concept for classification to prevent misinterpretation. In this study, we propose a method using a deep learning model to learn the image concept and then using the Knockoff samples to select the important concepts for prediction by controlling the False Discovery Rate (FDR) under a certain value. We evaluate the proposed method in our synthetic and real data experiments. Also, it shows that our method can control the FDR properly while selecting highly interpretable concepts to improve the trustworthiness of the model.
comment: Accepted to IJCAI'23
A Generalization of ViT/MLP-Mixer to Graphs ICML 2023
Graph Neural Networks (GNNs) have shown great potential in the field of graph representation learning. Standard GNNs define a local message-passing mechanism which propagates information over the whole graph domain by stacking multiple layers. This paradigm suffers from two major limitations, over-squashing and poor long-range dependencies, that can be solved using global attention but significantly increases the computational cost to quadratic complexity. In this work, we propose an alternative approach to overcome these structural limitations by leveraging the ViT/MLP-Mixer architectures introduced in computer vision. We introduce a new class of GNNs, called Graph ViT/MLP-Mixer, that holds three key properties. First, they capture long-range dependency and mitigate the issue of over-squashing as demonstrated on Long Range Graph Benchmark and TreeNeighbourMatch datasets. Second, they offer better speed and memory efficiency with a complexity linear to the number of nodes and edges, surpassing the related Graph Transformer and expressive GNN models. Third, they show high expressivity in terms of graph isomorphism as they can distinguish at least 3-WL non-isomorphic graphs. We test our architecture on 4 simulated datasets and 7 real-world benchmarks, and show highly competitive results on all of them. The source code is available for reproducibility at: \url{https://github.com/XiaoxinHe/Graph-ViT-MLPMixer}.
comment: In Proceedings of ICML 2023
♻ ☆ Open-VCLIP: Transforming CLIP to an Open-vocabulary Video Model via Interpolated Weight Optimization ICML 2023
Contrastive Language-Image Pretraining (CLIP) has demonstrated impressive zero-shot learning abilities for image understanding, yet limited effort has been made to investigate CLIP for zero-shot video recognition. We introduce Open-VCLIP, a simple yet effective approach that transforms CLIP into a strong zero-shot video classifier that can recognize unseen actions and events at test time. Our framework extends CLIP with minimal modifications to model spatial-temporal relationships in videos, making it a specialized video classifier, while striving for generalization. We formally show that training an Open-VCLIP is equivalent to continual learning with zero historical data. To address this problem, we propose Interpolated Weight Optimization, which utilizes the benefit of weight interpolation in both training and test time. We evaluate our method on three popular and challenging action recognition datasets following various zero-shot evaluation protocols and we demonstrate our approach outperforms state-of-the-art methods by clear margins. In particular, we achieve 87.9%, 58.3%, 81.1% zero-shot accuracy on UCF, HMDB and Kinetics-600 respectively, outperforming state-of-the-art methods by 8.3%, 7.8% and 12.2%. Code is released at https://github.com/wengzejia1/Open-VCLIP.
comment: 12 pages, 4 figures, ICML 2023
♻ ☆ StyleAvatar3D: Leveraging Image-Text Diffusion Models for High-Fidelity 3D Avatar Generation
The recent advancements in image-text diffusion models have stimulated research interest in large-scale 3D generative models. Nevertheless, the limited availability of diverse 3D resources presents significant challenges to learning. In this paper, we present a novel method for generating high-quality, stylized 3D avatars that utilizes pre-trained image-text diffusion models for data generation and a Generative Adversarial Network (GAN)-based 3D generation network for training. Our method leverages the comprehensive priors of appearance and geometry offered by image-text diffusion models to generate multi-view images of avatars in various styles. During data generation, we employ poses extracted from existing 3D models to guide the generation of multi-view images. To address the misalignment between poses and images in data, we investigate view-specific prompts and develop a coarse-to-fine discriminator for GAN training. We also delve into attribute-related prompts to increase the diversity of the generated avatars. Additionally, we develop a latent diffusion model within the style space of StyleGAN to enable the generation of avatars based on image inputs. Our approach demonstrates superior performance over current state-of-the-art methods in terms of visual quality and diversity of the produced avatars.
comment: Project page: https://github.com/icoz69/StyleAvatar3D
♻ ☆ 3DGTN: 3D Dual-Attention GLocal Transformer Network for Point Cloud Classification and Segmentation
Although the application of Transformers in 3D point cloud processing has achieved significant progress and success, it is still challenging for existing 3D Transformer methods to efficiently and accurately learn both valuable global features and valuable local features for improved applications. This paper presents a novel point cloud representational learning network, called 3D Dual Self-attention Global Local (GLocal) Transformer Network (3DGTN), for improved feature learning in both classification and segmentation tasks, with the following key contributions. First, a GLocal Feature Learning (GFL) block with the dual self-attention mechanism (i.e., a novel Point-Patch Self-Attention, called PPSA, and a channel-wise self-attention) is designed to efficiently learn the GLocal context information. Second, the GFL block is integrated with a multi-scale Graph Convolution-based Local Feature Aggregation (LFA) block, leading to a Global-Local (GLocal) information extraction module that can efficiently capture critical information. Third, a series of GLocal modules are used to construct a new hierarchical encoder-decoder structure to enable the learning of "GLocal" information in different scales in a hierarchical manner. The proposed framework is evaluated on both classification and segmentation datasets, demonstrating that the proposed method is capable of outperforming many state-of-the-art methods on both classification and segmentation tasks.
comment: 10 pages, 6 figures, 4 tables
♻ ☆ Uni-ControlNet: All-in-One Control to Text-to-Image Diffusion Models
Text-to-Image diffusion models have made tremendous progress over the past two years, enabling the generation of highly realistic images based on open-domain text descriptions. However, despite their success, text descriptions often struggle to adequately convey detailed controls, even when composed of long and complex texts. Moreover, recent studies have also shown that these models face challenges in understanding such complex texts and generating the corresponding images. Therefore, there is a growing need to enable more control modes beyond text description. In this paper, we introduce Uni-ControlNet, a novel approach that allows for the simultaneous utilization of different local controls (e.g., edge maps, depth map, segmentation masks) and global controls (e.g., CLIP image embeddings) in a flexible and composable manner within one model. Unlike existing methods, Uni-ControlNet only requires the fine-tuning of two additional adapters upon frozen pre-trained text-to-image diffusion models, eliminating the huge cost of training from scratch. Moreover, thanks to some dedicated adapter designs, Uni-ControlNet only necessitates a constant number (i.e., 2) of adapters, regardless of the number of local or global controls used. This not only reduces the fine-tuning costs and model size, making it more suitable for real-world deployment, but also facilitate composability of different conditions. Through both quantitative and qualitative comparisons, Uni-ControlNet demonstrates its superiority over existing methods in terms of controllability, generation quality and composability. Code is available at \url{https://github.com/ShihaoZhaoZSH/Uni-ControlNet}.
comment: Code is available at https://github.com/ShihaoZhaoZSH/Uni-ControlNet
♻ ☆ BRIGHT: Bi-level Feature Representation of Image Collections using Groups of Hash Tables
We present BRIGHT, a bi-level feature representation for an image collection, consisting of a per-image latent space on top of a multi-scale feature grid space. Our representation is learned by an autoencoder to encode images into continuous key codes, which are used to retrieve features from groups of multi-resolution hash tables. Our key codes and hash tables are trained together continuously with well-defined gradient flows, leading to high usage of the hash table entries and improved generative modeling compared to discrete Vector Quantization (VQ). Differently from existing continuous representations such as KL-regularized latent codes, our key codes are strictly bounded in scale and variance. Overall, feature encoding by BRIGHT is compact, efficient to train, and enables generative modeling over the image codes using state-of-the-art generators such as latent diffusion models(LDMs). Experimental results show that our method achieves comparable reconstruction results to VQ methods while having a smaller and more efficient decoder network. By applying LDM over our key code space, we achieve state-of-the-art performance on image synthesis on the LSUN-Church and human-face datasets.
comment: project page: https://bright-project01.github.io/
♻ ☆ Human-Machine Comparison for Cross-Race Face Verification: Race Bias at the Upper Limits of Performance?
Face recognition algorithms perform more accurately than humans in some cases, though humans and machines both show race-based accuracy differences. As algorithms continue to improve, it is important to continually assess their race bias relative to humans. We constructed a challenging test of 'cross-race' face verification and used it to compare humans and two state-of-the-art face recognition systems. Pairs of same- and different-identity faces of White and Black individuals were selected to be difficult for humans and an open-source implementation of the ArcFace face recognition algorithm from 2019 (5). Human participants (54 Black; 51 White) judged whether face pairs showed the same identity or different identities on a 7-point Likert-type scale. Two top-performing face recognition systems from the Face Recognition Vendor Test-ongoing performed the same test (7). By design, the test proved challenging for humans as a group, who performed above chance, but far less than perfect. Both state-of-the-art face recognition systems scored perfectly (no errors), consequently with equal accuracy for both races. We conclude that state-of-the-art systems for identity verification between two frontal face images of Black and White individuals can surpass the general population. Whether this result generalizes to challenging in-the-wild images is a pressing concern for deploying face recognition systems in unconstrained environments.
comment: 8 pages, 6 figures
♻ ☆ Uncertainty in Real-Time Semantic Segmentation on Embedded Systems CVPR 2023
Application for semantic segmentation models in areas such as autonomous vehicles and human computer interaction require real-time predictive capabilities. The challenges of addressing real-time application is amplified by the need to operate on resource constrained hardware. Whilst development of real-time methods for these platforms has increased, these models are unable to sufficiently reason about uncertainty present. This paper addresses this by combining deep feature extraction from pre-trained models with Bayesian regression and moment propagation for uncertainty aware predictions. We demonstrate how the proposed method can yield meaningful uncertainty on embedded hardware in real-time whilst maintaining predictive performance.
comment: Accepted CVPR 2023 Embedded Vision Workshop
♻ ☆ Error Estimation for Single-Image Human Body Mesh Reconstruction
Human pose and shape estimation methods continue to suffer in situations where one or more parts of the body are occluded. More importantly, these methods cannot express when their predicted pose is incorrect. This has serious consequences when these methods are used in human-robot interaction scenarios, where we need methods that can evaluate their predictions and flag situations where they might be wrong. This work studies this problem. We propose a method that combines information from OpenPose and SPIN -- two popular human pose and shape estimation methods -- to highlight regions on the predicted mesh that are least reliable. We have evaluated the proposed approach on 3DPW, 3DOH, and Human3.6M datasets, and the results demonstrate our model's effectiveness in identifying inaccurate regions of the human body mesh. Our code is available at https://github.com/Hamoon1987/meshConfidence.
Information Retrieval 16
☆ Structure-Aware Language Model Pretraining Improves Dense Retrieval on Structured Data ACL 2023
This paper presents Structure Aware Dense Retrieval (SANTA) model, which encodes user queries and structured data in one universal embedding space for retrieving structured data. SANTA proposes two pretraining methods to make language models structure-aware and learn effective representations for structured data: 1) Structured Data Alignment, which utilizes the natural alignment relations between structured data and unstructured data for structure-aware pretraining. It contrastively trains language models to represent multi-modal text data and teaches models to distinguish matched structured data for unstructured texts. 2) Masked Entity Prediction, which designs an entity-oriented mask strategy and asks language models to fill in the masked entities. Our experiments show that SANTA achieves state-of-the-art on code search and product search and conducts convincing results in the zero-shot setting. SANTA learns tailored representations for multi-modal text data by aligning structured and unstructured data pairs and capturing structural semantics by masking and predicting entities in the structured data. All codes are available at https://github.com/OpenMatch/OpenMatch.
comment: Accepted by Findings of ACL 2023
☆ Web scraping: a promising tool for geographic data acquisition
With much of our lives taking place online, researchers are increasingly turning to information from the World Wide Web to gain insights into geographic patterns and processes. Web scraping as an online data acquisition technique allows us to gather intelligence especially on social and economic actions for which the Web serves as a platform. Specific opportunities relate to near-real-time access to object-level geolocated data, which can be captured in a cost-effective way. The studied geographic phenomena include, but are not limited to, the rental market and associated processes such as gentrification, entrepreneurial ecosystems, or spatial planning processes. Since the information retrieved from the Web is not made available for that purpose, Web scraping faces several unique challenges, several of which relate to location. Ethical and legal issues mainly relate to intellectual property rights, informed consent and (geo-) privacy, and website integrity and contract. These issues also effect the practice of open science. In addition, there are technical and statistical challenges that relate to dependability and incompleteness, data inconsistencies and bias, as well as the limited historical coverage. Geospatial analyses furthermore usually require the automated extraction and subsequent resolution of toponyms or addresses (geoparsing, geocoding). A study on apartment rent in Leipzig, Germany is used to illustrate the use of Web scraping and its challenges. We conclude that geographic researchers should embrace Web scraping as a powerful and affordable digital fieldwork tool while paying special attention to its legal, ethical, and methodological challenges.
comment: 18 pages, 7 figures, 2 tables
☆ A Survey on Large Language Models for Recommendation
Large Language Models (LLMs) have emerged as powerful tools in the field of Natural Language Processing (NLP) and have recently gained significant attention in the domain of Recommendation Systems (RS). These models, trained on massive amounts of data using self-supervised learning, have demonstrated remarkable success in learning universal representations and have the potential to enhance various aspects of recommendation systems by some effective transfer techniques such as fine-tuning and prompt tuning, and so on. The crucial aspect of harnessing the power of language models in enhancing recommendation quality is the utilization of their high-quality representations of textual features and their extensive coverage of external knowledge to establish correlations between items and users. To provide a comprehensive understanding of the existing LLM-based recommendation systems, this survey presents a taxonomy that categorizes these models into two major paradigms, respectively Discriminative LLM for Recommendation (DLLM4Rec) and Generative LLM for Recommendation (GLLM4Rec), with the latter being systematically sorted out for the first time. Furthermore, we systematically review and analyze existing LLM-based recommendation systems within each paradigm, providing insights into their methodologies, techniques, and performance. Additionally, we identify key challenges and several valuable findings to provide researchers and practitioners with inspiration.
comment: 10 pages, 3 figures
☆ BEIR-PL: Zero Shot Information Retrieval Benchmark for the Polish Language
The BEIR dataset is a large, heterogeneous benchmark for Information Retrieval (IR) in zero-shot settings, garnering considerable attention within the research community. However, BEIR and analogous datasets are predominantly restricted to the English language. Our objective is to establish extensive large-scale resources for IR in the Polish language, thereby advancing the research in this NLP area. In this work, inspired by mMARCO and Mr.~TyDi datasets, we translated all accessible open IR datasets into Polish, and we introduced the BEIR-PL benchmark -- a new benchmark which comprises 13 datasets, facilitating further development, training and evaluation of modern Polish language models for IR tasks. We executed an evaluation and comparison of numerous IR models on the newly introduced BEIR-PL benchmark. Furthermore, we publish pre-trained open IR models for Polish language,d marking a pioneering development in this field. Additionally, the evaluation revealed that BM25 achieved significantly lower scores for Polish than for English, which can be attributed to high inflection and intricate morphological structure of the Polish language. Finally, we trained various re-ranking models to enhance the BM25 retrieval, and we compared their performance to identify their unique characteristic features. To ensure accurate model comparisons, it is necessary to scrutinise individual results rather than to average across the entire benchmark. Thus, we thoroughly analysed the outcomes of IR models in relation to each individual data subset encompassed by the BEIR benchmark. The benchmark data is available at URL {\bf https://huggingface.co/clarin-knext}.
☆ Medication Recommendation via Domain Knowledge Informed Deep Learning
Medication recommendation is a fundamental yet crucial branch of healthcare, which provides opportunities to support clinical physicians with more accurate medication prescriptions for patients with complex health conditions. Learning from electronic health records (EHR) to recommend medications is the most common way in previous studies. However, most of them neglect incorporating domain knowledge according to the clinical manifestations in the EHR of the patient. To address these issues, we propose a novel \textbf{D}omain \textbf{K}nowledge \textbf{I}nformed \textbf{Net}work (DKINet) to integrate domain knowledge with observable clinical manifestations of the patient, which is the first dynamic domain knowledge informed framework toward medication recommendation. In particular, we first design a knowledge-driven encoder to capture the domain information and then develop a data-driven encoder to integrate domain knowledge into the observable EHR. To endow the model with the capability of temporal decision, we design an explicit medication encoder for learning the longitudinal dependence of the patient. Extensive experiments on three publicly available datasets verify the superiority of our method. The code will be public upon acceptance.
comment: 11 pages, 4 figures
☆ Towards Semi-supervised Universal Graph Classification
Graph neural networks have pushed state-of-the-arts in graph classifications recently. Typically, these methods are studied within the context of supervised end-to-end training, which necessities copious task-specific labels. However, in real-world circumstances, labeled data could be limited, and there could be a massive corpus of unlabeled data, even from unknown classes as a complementary. Towards this end, we study the problem of semi-supervised universal graph classification, which not only identifies graph samples which do not belong to known classes, but also classifies the remaining samples into their respective classes. This problem is challenging due to a severe lack of labels and potential class shifts. In this paper, we propose a novel graph neural network framework named UGNN, which makes the best of unlabeled data from the subgraph perspective. To tackle class shifts, we estimate the certainty of unlabeled graphs using multiple subgraphs, which facilities the discovery of unlabeled data from unknown categories. Moreover, we construct semantic prototypes in the embedding space for both known and unknown categories and utilize posterior prototype assignments inferred from the Sinkhorn-Knopp algorithm to learn from abundant unlabeled graphs across different subgraph views. Extensive experiments on six datasets verify the effectiveness of UGNN in different settings.
comment: Accepted by IEEE Transactions on Knowledge and Data Engineering (TKDE 2023)
☆ Multi-Epoch Learning for Deep Click-Through Rate Prediction Models
The one-epoch overfitting phenomenon has been widely observed in industrial Click-Through Rate (CTR) applications, where the model performance experiences a significant degradation at the beginning of the second epoch. Recent advances try to understand the underlying factors behind this phenomenon through extensive experiments. However, it is still unknown whether a multi-epoch training paradigm could achieve better results, as the best performance is usually achieved by one-epoch training. In this paper, we hypothesize that the emergence of this phenomenon may be attributed to the susceptibility of the embedding layer to overfitting, which can stem from the high-dimensional sparsity of data. To maintain feature sparsity while simultaneously avoiding overfitting of embeddings, we propose a novel Multi-Epoch learning with Data Augmentation (MEDA), which can be directly applied to most deep CTR models. MEDA achieves data augmentation by reinitializing the embedding layer in each epoch, thereby avoiding embedding overfitting and simultaneously improving convergence. To our best knowledge, MEDA is the first multi-epoch training paradigm designed for deep CTR prediction models. We conduct extensive experiments on several public datasets, and the effectiveness of our proposed MEDA is fully verified. Notably, the results show that MEDA can significantly outperform the conventional one-epoch training. Besides, MEDA has exhibited significant benefits in a real-world scene on Kuaishou.
☆ TransAct: Transformer-based Realtime User Action Model for Recommendation at Pinterest KDD'23
Sequential models that encode user activity for next action prediction have become a popular design choice for building web-scale personalized recommendation systems. Traditional methods of sequential recommendation either utilize end-to-end learning on realtime user actions, or learn user representations separately in an offline batch-generated manner. This paper (1) presents Pinterest's ranking architecture for Homefeed, our personalized recommendation product and the largest engagement surface; (2) proposes TransAct, a sequential model that extracts users' short-term preferences from their realtime activities; (3) describes our hybrid approach to ranking, which combines end-to-end sequential modeling via TransAct with batch-generated user embeddings. The hybrid approach allows us to combine the advantages of responsiveness from learning directly on realtime user activity with the cost-effectiveness of batch user representations learned over a longer time period. We describe the results of ablation studies, the challenges we faced during productionization, and the outcome of an online A/B experiment, which validates the effectiveness of our hybrid ranking model. We further demonstrate the effectiveness of TransAct on other surfaces such as contextual recommendations and search. Our model has been deployed to production in Homefeed, Related Pins, Notifications, and Search at Pinterest.
comment: \c{opyright} {ACM} {2023}. This is the author's version of the work. It is posted here for your personal use. Not for redistribution. The definitive Version of Record was published in KDD'23, http://dx.doi.org/10.1145/3580305.3599918
☆ VMap: An Interactive Rectangular Space-filling Visualization for Map-like Vertex-centric Graph Exploration IEEE VIS
We present VMap, a map-like rectangular space-filling visualization, to perform vertex-centric graph exploration. Existing visualizations have limited support for quality optimization among rectangular aspect ratios, vertex-edge intersection, and data encoding accuracy. To tackle this problem, VMap integrates three novel components: (1) a desired-aspect-ratio (DAR) rectangular partitioning algorithm, (2) a two-stage rectangle adjustment algorithm, and (3) a simulated annealing based heuristic optimizer. First, to generate a rectangular space-filling layout of an input graph, we subdivide the 2D embedding of the graph into rectangles with optimization of rectangles' aspect ratios toward a desired aspect ratio. Second, to route graph edges between rectangles without vertex-edge occlusion, we devise a two-stage algorithm to adjust a rectangular layout to insert border space between rectangles. Third, to produce and arrange rectangles by considering multiple visual criteria, we design a simulated annealing based heuristic optimization to adjust vertices' 2D embedding to support trade-offs among aspect ratio quality and the encoding accuracy of vertices' weights and adjacency. We evaluated the effectiveness of VMap on both synthetic and application datasets. The resulting rectangular layout has better aspect ratio quality on synthetic data compared with the existing method for the rectangular partitioning of 2D points. On three real-world datasets, VMap achieved better encoding accuracy and attained faster generation speed compared with existing methods on graphs' rectangular layout generation. We further illustrate the usefulness of VMap for vertex-centric graph exploration through three case studies on visualizing social networks, representing academic communities, and displaying geographic information.
comment: Submitted to IEEE Visualization Conference (IEEE VIS) 2019 and 2022
☆ Beyond Rankings: Exploring the Impact of SERP Features on Organic Click-through Rates
Search Engine Result Pages (SERPs) serve as the digital gateways to the vast expanse of the internet. Past decades have witnessed a surge in research primarily centered on the influence of website ranking on these pages, to determine the click-through rate (CTR). However, during this period, the landscape of SERPs has undergone a dramatic evolution: SERP features, encompassing elements such as knowledge panels, media galleries, FAQs, and more, have emerged as an increasingly prominent facet of these result pages. Our study examines the crucial role of these features, revealing them to be not merely aesthetic components, but strongly influence CTR and the associated behavior of internet users. We demonstrate how these features can significantly modulate web traffic, either amplifying or attenuating it. We dissect these intricate interaction effects leveraging a unique dataset of 67,000 keywords and their respective Google SERPs, spanning over 40 distinct US-based e-commerce domains, generating over 6 million clicks from 24 million views. This cross-website dataset, unprecedented in its scope, enables us to assess the impact of 24 different SERP features on organic CTR. Through an ablation study modeling CTR, we illustrate the incremental predictive power these features hold.
comment: submitted IEEE DSAA conference, 14 pages, 5 figures, 2 tables
☆ The Effect of News Article Quality on Ad Consumption CIKM2021
Practical news feed platforms generate a hybrid list of news articles and advertising items (e.g., products, services, or information) and many platforms optimize the position of news articles and advertisements independently. However, they should be arranged with careful consideration of each other, as we show in this study, since user behaviors toward advertisements are significantly affected by the news articles. This paper investigates the effect of news articles on users' ad consumption and shows the dependency between news and ad effectiveness. We conducted a service log analysis and showed that sessions with high-quality news article exposure had more ad consumption than those with low-quality news article exposure. Based on this result, we hypothesized that exposure to high-quality articles will lead to a high ad consumption rate. Thus, we conducted million-scale A/B testing to investigate the effect of high-quality articles on ad consumption, in which we prioritized high-quality articles in the ranking for the treatment group. The A/B test showed that the treatment group's ad consumption, such as the number of clicks, conversions, and sales, increased significantly while the number of article clicks decreased. We also found that users who prefer a social or economic topic had more ad consumption by stratified analysis. These insights regarding news articles and advertisements will help optimize news and ad effectiveness in rankings considering their mutual influence.
comment: 30th ACM International Conference on Information and Knowledge Management (CIKM2021)
♻ ☆ RARR: Researching and Revising What Language Models Say, Using Language Models ACL 2023
Language models (LMs) now excel at many tasks such as few-shot learning, question answering, reasoning, and dialog. However, they sometimes generate unsupported or misleading content. A user cannot easily determine whether their outputs are trustworthy or not, because most LMs do not have any built-in mechanism for attribution to external evidence. To enable attribution while still preserving all the powerful advantages of recent generation models, we propose RARR (Retrofit Attribution using Research and Revision), a system that 1) automatically finds attribution for the output of any text generation model and 2) post-edits the output to fix unsupported content while preserving the original output as much as possible. When applied to the output of several state-of-the-art LMs on a diverse set of generation tasks, we find that RARR significantly improves attribution while otherwise preserving the original input to a much greater degree than previously explored edit models. Furthermore, the implementation of RARR requires only a handful of training examples, a large language model, and standard web search.
comment: ACL 2023
♻ ☆ Understanding Diversity in Session-Based Recommendation
Current session-based recommender systems (SBRSs) mainly focus on maximizing recommendation accuracy, while few studies have been devoted to improve diversity beyond accuracy. Meanwhile, it is unclear how the accuracy-oriented SBRSs perform in terms of diversity. Besides, the asserted "trade-off" relationship between accuracy and diversity has been increasingly questioned in the literature. Towards the aforementioned issues, we conduct a holistic study to particularly examine the recommendation performance of representative SBRSs w.r.t. both accuracy and diversity, striving for better understanding the diversity-related issues for SBRSs and providing guidance on designing diversified SBRSs. Particularly, for a fair and thorough comparison, we deliberately select state-of-the-art non-neural, deep neural, and diversified SBRSs, by covering more scenarios with appropriate experimental setups, e.g., representative datasets, evaluation metrics, and hyper-parameter optimization technique. Our empirical results unveil that: 1) non-diversified methods can also obtain satisfying performance on diversity, which might even surpass diversified ones; and 2) the relationship between accuracy and diversity is quite complex. Besides the "trade-off" relationship, they might be positively correlated with each other, that is, having a same-trend (win-win or lose-lose) relationship, which varies across different methods and datasets. Additionally, we further identify three possible influential factors on diversity in SBRSs (i.e., granularity of item categorization, session diversity of datasets, and length of recommendation lists).
♻ ☆ Variational Open-Domain Question Answering ICML 2023
Retrieval-augmented models have proven to be effective in natural language processing tasks, yet there remains a lack of research on their optimization using variational inference. We introduce the Variational Open-Domain (VOD) framework for end-to-end training and evaluation of retrieval-augmented models, focusing on open-domain question answering and language modelling. The VOD objective, a self-normalized estimate of the R\'enyi variational bound, approximates the task marginal likelihood and is evaluated under samples drawn from an auxiliary sampling distribution (cached retriever and/or approximate posterior). It remains tractable, even for retriever distributions defined on large corpora. We demonstrate VOD's versatility by training reader-retriever BERT-sized models on multiple-choice medical exam questions. On the MedMCQA dataset, we outperform the domain-tuned Med-PaLM by +5.3% despite using 2.500$\times$ fewer parameters. Our retrieval-augmented BioLinkBERT model scored 62.9% on the MedMCQA and 55.0% on the MedQA-USMLE. Last, we show the effectiveness of our learned retriever component in the context of medical semantic search.
comment: 28 pages, 5 figures. Accepted at ICML 2023
♻ ☆ Unified Embedding: Battle-Tested Feature Representations for Web-Scale ML Systems
Learning high-quality feature embeddings efficiently and effectively is critical for the performance of web-scale machine learning systems. A typical model ingests hundreds of features with vocabularies on the order of millions to billions of tokens. The standard approach is to represent each feature value as a d-dimensional embedding, introducing hundreds of billions of parameters for extremely high-cardinality features. This bottleneck has led to substantial progress in alternative embedding algorithms. Many of these methods, however, make the assumption that each feature uses an independent embedding table. This work introduces a simple yet highly effective framework, Feature Multiplexing, where one single representation space is used across many different categorical features. Our theoretical and empirical analysis reveals that multiplexed embeddings can be decomposed into components from each constituent feature, allowing models to distinguish between features. We show that multiplexed representations lead to Pareto-optimal parameter-accuracy tradeoffs for three public benchmark datasets. Further, we propose a highly practical approach called Unified Embedding with three major benefits: simplified feature configuration, strong adaptation to dynamic data distributions, and compatibility with modern hardware. Unified embedding gives significant improvements in offline and online metrics compared to highly competitive baselines across five web-scale search, ads, and recommender systems, where it serves billions of users across the world in industry-leading products.
♻ ☆ Reasoning with Language Model Prompting: A Survey ACL 2023
Reasoning, as an essential ability for complex problem-solving, can provide back-end support for various real-world applications, such as medical diagnosis, negotiation, etc. This paper provides a comprehensive survey of cutting-edge research on reasoning with language model prompting. We introduce research works with comparisons and summaries and provide systematic resources to help beginners. We also discuss the potential reasons for emerging such reasoning abilities and highlight future research directions. Resources are available at https://github.com/zjunlp/Prompt4ReasoningPapers (updated periodically).
comment: ACL 2023
Machine Learning 150
☆ Improving CLIP Training with Language Rewrites
Contrastive Language-Image Pre-training (CLIP) stands as one of the most effective and scalable methods for training transferable vision models using paired image and text data. CLIP models are trained using contrastive loss, which typically relies on data augmentations to prevent overfitting and shortcuts. However, in the CLIP training paradigm, data augmentations are exclusively applied to image inputs, while language inputs remain unchanged throughout the entire training process, limiting the exposure of diverse texts to the same image. In this paper, we introduce Language augmented CLIP (LaCLIP), a simple yet highly effective approach to enhance CLIP training through language rewrites. Leveraging the in-context learning capability of large language models, we rewrite the text descriptions associated with each image. These rewritten texts exhibit diversity in sentence structure and vocabulary while preserving the original key concepts and meanings. During training, LaCLIP randomly selects either the original texts or the rewritten versions as text augmentations for each image. Extensive experiments on CC3M, CC12M, RedCaps and LAION-400M datasets show that CLIP pre-training with language rewrites significantly improves the transfer performance without computation or memory overhead during training. Specifically for ImageNet zero-shot accuracy, LaCLIP outperforms CLIP by 8.2% on CC12M and 2.4% on LAION-400M. Code is available at https://github.com/LijieFan/LaCLIP.
☆ Understanding and Mitigating Copying in Diffusion Models
Images generated by diffusion models like Stable Diffusion are increasingly widespread. Recent works and even lawsuits have shown that these models are prone to replicating their training data, unbeknownst to the user. In this paper, we first analyze this memorization problem in text-to-image diffusion models. While it is widely believed that duplicated images in the training set are responsible for content replication at inference time, we observe that the text conditioning of the model plays a similarly important role. In fact, we see in our experiments that data replication often does not happen for unconditional models, while it is common in the text-conditional case. Motivated by our findings, we then propose several techniques for reducing data replication at both training and inference time by randomizing and augmenting image captions in the training set.
comment: 17 pages, preprint. Code is available at https://github.com/somepago/DCR
☆ Efficient Diffusion Policies for Offline Reinforcement Learning
Offline reinforcement learning (RL) aims to learn optimal policies from offline datasets, where the parameterization of policies is crucial but often overlooked. Recently, Diffsuion-QL significantly boosts the performance of offline RL by representing a policy with a diffusion model, whose success relies on a parametrized Markov Chain with hundreds of steps for sampling. However, Diffusion-QL suffers from two critical limitations. 1) It is computationally inefficient to forward and backward through the whole Markov chain during training. 2) It is incompatible with maximum likelihood-based RL algorithms (e.g., policy gradient methods) as the likelihood of diffusion models is intractable. Therefore, we propose efficient diffusion policy (EDP) to overcome these two challenges. EDP approximately constructs actions from corrupted ones at training to avoid running the sampling chain. We conduct extensive experiments on the D4RL benchmark. The results show that EDP can reduce the diffusion policy training time from 5 days to 5 hours on gym-locomotion tasks. Moreover, we show that EDP is compatible with various offline RL algorithms (TD3, CRR, and IQL) and achieves new state-of-the-art on D4RL by large margins over previous methods. Our code is available at https://github.com/sail-sg/edp.
comment: preprint
☆ Managed Geo-Distributed Feature Store: Architecture and System Design
Companies are using machine learning to solve real-world problems and are developing hundreds to thousands of features in the process. They are building feature engineering pipelines as part of MLOps life cycle to transform data from various data sources and materialize the same for future consumption. Without feature stores, different teams across various business groups would maintain the above process independently, which can lead to conflicting and duplicated features in the system. Data scientists find it hard to search for and reuse existing features and it is painful to maintain version control. Furthermore, feature correctness violations related to online (inferencing) - offline (training) skews and data leakage are common. Although the machine learning community has extensively discussed the need for feature stores and their purpose, this paper aims to capture the core architectural components that make up a managed feature store and to share the design learning in building such a system.
comment: All the authors are from the AzureML Feature Store product group and are listed in alphabetical order. Bhala Ranganathan: System architect and tech lead of AzureML Feature Store. Feng Pan, Qianjun Xu: Engineering managers. Sethu Raman: Product Manager of AzureML Feature Store who structured and organized the product vision and specifications
☆ Feature Learning in Image Hierarchies using Functional Maximal Correlation
This paper proposes the Hierarchical Functional Maximal Correlation Algorithm (HFMCA), a hierarchical methodology that characterizes dependencies across two hierarchical levels in multiview systems. By framing view similarities as dependencies and ensuring contrastivity by imposing orthonormality, HFMCA achieves faster convergence and increased stability in self-supervised learning. HFMCA defines and measures dependencies within image hierarchies, from pixels and patches to full images. We find that the network topology for approximating orthonormal basis functions aligns with a vanilla CNN, enabling the decomposition of density ratios between neighboring layers of feature maps. This approach provides powerful interpretability, revealing the resemblance between supervision and self-supervision through the lens of internal representations.
☆ Alternating Minimization for Regression with Tropical Rational Functions
We propose an alternating minimization heuristic for regression over the space of tropical rational functions with fixed exponents. The method alternates between fitting the numerator and denominator terms via tropical polynomial regression, which is known to admit a closed form solution. We demonstrate the behavior of the alternating minimization method experimentally. Experiments demonstrate that the heuristic provides a reasonable approximation of the input data. Our work is motivated by applications to ReLU neural networks, a popular class of network architectures in the machine learning community which are closely related to tropical rational functions.
☆ A survey on the complexity of learning quantum states
We survey various recent results that rigorously study the complexity of learning quantum states. These include progress on quantum tomography, learning physical quantum states, alternate learning models to tomography and learning classical functions encoded as quantum states. We highlight how these results are paving the way for a highly successful theory with a range of exciting open questions. To this end, we distill 25 open questions from these results.
comment: Invited article by Nature Review Physics. 39 pages, 6 figures
☆ TOFG: A Unified and Fine-Grained Environment Representation in Autonomous Driving ICRA 2023
In autonomous driving, an accurate understanding of environment, e.g., the vehicle-to-vehicle and vehicle-to-lane interactions, plays a critical role in many driving tasks such as trajectory prediction and motion planning. Environment information comes from high-definition (HD) map and historical trajectories of vehicles. Due to the heterogeneity of the map data and trajectory data, many data-driven models for trajectory prediction and motion planning extract vehicle-to-vehicle and vehicle-to-lane interactions in a separate and sequential manner. However, such a manner may capture biased interpretation of interactions, causing lower prediction and planning accuracy. Moreover, separate extraction leads to a complicated model structure and hence the overall efficiency and scalability are sacrificed. To address the above issues, we propose an environment representation, Temporal Occupancy Flow Graph (TOFG). Specifically, the occupancy flow-based representation unifies the map information and vehicle trajectories into a homogeneous data format and enables a consistent prediction. The temporal dependencies among vehicles can help capture the change of occupancy flow timely to further promote model performance. To demonstrate that TOFG is capable of simplifying the model architecture, we incorporate TOFG with a simple graph attention (GAT) based neural network and propose TOFG-GAT, which can be used for both trajectory prediction and motion planning. Experiment results show that TOFG-GAT achieves better or competitive performance than all the SOTA baselines with less training time.
comment: Accepted by ICRA 2023
☆ Latent Exploration for Reinforcement Learning
In Reinforcement Learning, agents learn policies by exploring and interacting with the environment. Due to the curse of dimensionality, learning policies that map high-dimensional sensory input to motor output is particularly challenging. During training, state of the art methods (SAC, PPO, etc.) explore the environment by perturbing the actuation with independent Gaussian noise. While this unstructured exploration has proven successful in numerous tasks, it ought to be suboptimal for overactuated systems. When multiple actuators, such as motors or muscles, drive behavior, uncorrelated perturbations risk diminishing each other's effect, or modifying the behavior in a task-irrelevant way. While solutions to introduce time correlation across action perturbations exist, introducing correlation across actuators has been largely ignored. Here, we propose LATent TIme-Correlated Exploration (Lattice), a method to inject temporally-correlated noise into the latent state of the policy network, which can be seamlessly integrated with on- and off-policy algorithms. We demonstrate that the noisy actions generated by perturbing the network's activations can be modeled as a multivariate Gaussian distribution with a full covariance matrix. In the PyBullet locomotion tasks, Lattice-SAC achieves state of the art results, and reaches 18% higher reward than unstructured exploration in the Humanoid environment. In the musculoskeletal control environments of MyoSuite, Lattice-PPO achieves higher reward in most reaching and object manipulation tasks, while also finding more energy-efficient policies with reductions of 20-60%. Overall, we demonstrate the effectiveness of structured action noise in time and actuator space for complex motor control tasks.
comment: Code available at https://github.com/amathislab/lattice
☆ Three-Way Trade-Off in Multi-Objective Learning: Optimization, Generalization and Conflict-Avoidance
Multi-objective learning (MOL) problems often arise in emerging machine learning problems when there are multiple learning criteria or multiple learning tasks. Recent works have developed various dynamic weighting algorithms for MOL such as MGDA and its variants, where the central idea is to find an update direction that avoids conflicts among objectives. Albeit its appealing intuition, empirical studies show that dynamic weighting methods may not always outperform static ones. To understand this theory-practical gap, we focus on a new stochastic variant of MGDA - the Multi-objective gradient with Double sampling (MoDo) algorithm, and study the generalization performance of the dynamic weighting-based MoDo and its interplay with optimization through the lens of algorithm stability. Perhaps surprisingly, we find that the key rationale behind MGDA -- updating along conflict-avoidant direction - may hinder dynamic weighting algorithms from achieving the optimal ${\cal O}(1/\sqrt{n})$ population risk, where $n$ is the number of training samples. We further demonstrate the variability of dynamic weights on the three-way trade-off among optimization, generalization, and conflict avoidance that is unique in MOL.
☆ Rare Life Event Detection via Mobile Sensing Using Multi-Task Learning
Rare life events significantly impact mental health, and their detection in behavioral studies is a crucial step towards health-based interventions. We envision that mobile sensing data can be used to detect these anomalies. However, the human-centered nature of the problem, combined with the infrequency and uniqueness of these events makes it challenging for unsupervised machine learning methods. In this paper, we first investigate granger-causality between life events and human behavior using sensing data. Next, we propose a multi-task framework with an unsupervised autoencoder to capture irregular behavior, and an auxiliary sequence predictor that identifies transitions in workplace performance to contextualize events. We perform experiments using data from a mobile sensing study comprising N=126 information workers from multiple industries, spanning 10106 days with 198 rare events (<2%). Through personalized inference, we detect the exact day of a rare event with an F1 of 0.34, demonstrating that our method outperforms several baselines. Finally, we discuss the implications of our work from the context of real-world deployment.
comment: 15 pages, 4 figures, CHIL 2023 (Accepted)
☆ UNSSOR: Unsupervised Neural Speech Separation by Leveraging Over-determined Training Mixtures
In reverberant conditions with multiple concurrent speakers, each microphone acquires a mixture signal of multiple speakers at a different location. In over-determined conditions where the microphones out-number speakers, we can narrow down the solutions to speaker images and realize unsupervised speech separation by leveraging each mixture signal as a constraint (i.e., the estimated speaker images at a microphone should add up to the mixture). Equipped with this insight, we propose UNSSOR, an algorithm for $\textbf{u}$nsupervised $\textbf{n}$eural $\textbf{s}$peech $\textbf{s}$eparation by leveraging $\textbf{o}$ver-determined training mixtu$\textbf{r}$es. At each training step, we feed an input mixture to a deep neural network (DNN) to produce an intermediate estimate for each speaker, linearly filter the estimates, and optimize a loss so that, at each microphone, the filtered estimates of all the speakers can add up to the mixture to satisfy the above constraint. We show that this loss can promote unsupervised separation of speakers. The linear filters are computed in each sub-band based on the mixture and DNN estimates through the forward convolutive prediction (FCP) algorithm. To address the frequency permutation problem incurred by using sub-band FCP, a loss term based on minimizing intra-source magnitude scattering is proposed. Although UNSSOR requires over-determined training mixtures, we can train DNNs to achieve under-determined separation (e.g., unsupervised monaural speech separation). Evaluation results on two-speaker separation in reverberant conditions show the effectiveness and potential of UNSSOR.
comment: in submission
☆ Efficient PDE-Constrained optimization under high-dimensional uncertainty using derivative-informed neural operators
We propose a novel machine learning framework for solving optimization problems governed by large-scale partial differential equations (PDEs) with high-dimensional random parameters. Such optimization under uncertainty (OUU) problems may be computational prohibitive using classical methods, particularly when a large number of samples is needed to evaluate risk measures at every iteration of an optimization algorithm, where each sample requires the solution of an expensive-to-solve PDE. To address this challenge, we propose a new neural operator approximation of the PDE solution operator that has the combined merits of (1) accurate approximation of not only the map from the joint inputs of random parameters and optimization variables to the PDE state, but also its derivative with respect to the optimization variables, (2) efficient construction of the neural network using reduced basis architectures that are scalable to high-dimensional OUU problems, and (3) requiring only a limited number of training data to achieve high accuracy for both the PDE solution and the OUU solution. We refer to such neural operators as multi-input reduced basis derivative informed neural operators (MR-DINOs). We demonstrate the accuracy and efficiency our approach through several numerical experiments, i.e. the risk-averse control of a semilinear elliptic PDE and the steady state Navier--Stokes equations in two and three spatial dimensions, each involving random field inputs. Across the examples, MR-DINOs offer $10^{3}$--$10^{7} \times$ reductions in execution time, and are able to produce OUU solutions of comparable accuracies to those from standard PDE based solutions while being over $10 \times$ more cost-efficient after factoring in the cost of construction.
☆ Integrated Decision Gradients: Compute Your Attributions Where the Model Makes Its Decision NeurIPS 2023
Attribution algorithms are frequently employed to explain the decisions of neural network models. Integrated Gradients (IG) is an influential attribution method due to its strong axiomatic foundation. The algorithm is based on integrating the gradients along a path from a reference image to the input image. Unfortunately, it can be observed that gradients computed from regions where the output logit changes minimally along the path provide poor explanations for the model decision, which is called the saturation effect problem. In this paper, we propose an attribution algorithm called integrated decision gradients (IDG). The algorithm focuses on integrating gradients from the region of the path where the model makes its decision, i.e., the portion of the path where the output logit rapidly transitions from zero to its final value. This is practically realized by scaling each gradient by the derivative of the output logit with respect to the path. The algorithm thereby provides a principled solution to the saturation problem. Additionally, we minimize the errors within the Riemann sum approximation of the path integral by utilizing non-uniform subdivisions determined by adaptive sampling. In the evaluation on ImageNet, it is demonstrated that IDG outperforms IG, left-IG, guided IG, and adversarial gradient integration both qualitatively and quantitatively using standard insertion and deletion metrics across three common models.
comment: 18 pages, 8 figures, submitted to NeurIPS 2023, the full code implementation of the paper results is located at: https://github.com/chasewalker26/Integrated-Decision-Gradients
☆ Let's Verify Step by Step
In recent years, large language models have greatly improved in their ability to perform complex multi-step reasoning. However, even state-of-the-art models still regularly produce logical mistakes. To train more reliable models, we can turn either to outcome supervision, which provides feedback for a final result, or process supervision, which provides feedback for each intermediate reasoning step. Given the importance of training reliable models, and given the high cost of human feedback, it is important to carefully compare the both methods. Recent work has already begun this comparison, but many questions still remain. We conduct our own investigation, finding that process supervision significantly outperforms outcome supervision for training models to solve problems from the challenging MATH dataset. Our process-supervised model solves 78% of problems from a representative subset of the MATH test set. Additionally, we show that active learning significantly improves the efficacy of process supervision. To support related research, we also release PRM800K, the complete dataset of 800,000 step-level human feedback labels used to train our best reward model.
☆ ActiveAED: A Human in the Loop Improves Annotation Error Detection ACL 2023
Manually annotated datasets are crucial for training and evaluating Natural Language Processing models. However, recent work has discovered that even widely-used benchmark datasets contain a substantial number of erroneous annotations. This problem has been addressed with Annotation Error Detection (AED) models, which can flag such errors for human re-annotation. However, even though many of these AED methods assume a final curation step in which a human annotator decides whether the annotation is erroneous, they have been developed as static models without any human-in-the-loop component. In this work, we propose ActiveAED, an AED method that can detect errors more accurately by repeatedly querying a human for error corrections in its prediction loop. We evaluate ActiveAED on eight datasets spanning five different tasks and find that it leads to improvements over the state of the art on seven of them, with gains of up to six percentage points in average precision.
comment: Findings of ACL 2023
☆ Deception by Omission: Using Adversarial Missingness to Poison Causal Structure Learning
Inference of causal structures from observational data is a key component of causal machine learning; in practice, this data may be incompletely observed. Prior work has demonstrated that adversarial perturbations of completely observed training data may be used to force the learning of inaccurate causal structural models (SCMs). However, when the data can be audited for correctness (e.g., it is crytographically signed by its source), this adversarial mechanism is invalidated. This work introduces a novel attack methodology wherein the adversary deceptively omits a portion of the true training data to bias the learned causal structures in a desired manner. Theoretically sound attack mechanisms are derived for the case of arbitrary SCMs, and a sample-efficient learning-based heuristic is given for Gaussian SCMs. Experimental validation of these approaches on real and synthetic data sets demonstrates the effectiveness of adversarial missingness attacks at deceiving popular causal structure learning algorithms.
☆ Tree-Ring Watermarks: Fingerprints for Diffusion Images that are Invisible and Robust
Watermarking the outputs of generative models is a crucial technique for tracing copyright and preventing potential harm from AI-generated content. In this paper, we introduce a novel technique called Tree-Ring Watermarking that robustly fingerprints diffusion model outputs. Unlike existing methods that perform post-hoc modifications to images after sampling, Tree-Ring Watermarking subtly influences the entire sampling process, resulting in a model fingerprint that is invisible to humans. The watermark embeds a pattern into the initial noise vector used for sampling. These patterns are structured in Fourier space so that they are invariant to convolutions, crops, dilations, flips, and rotations. After image generation, the watermark signal is detected by inverting the diffusion process to retrieve the noise vector, which is then checked for the embedded signal. We demonstrate that this technique can be easily applied to arbitrary diffusion models, including text-conditioned Stable Diffusion, as a plug-in with negligible loss in FID. Our watermark is semantically hidden in the image space and is far more robust than watermarking alternatives that are currently deployed. Code is available at github.com/YuxinWenRick/tree-ring-watermark.
comment: 16 pages, 8 figures, code is available at https://github.com/YuxinWenRick/tree-ring-watermark
☆ A Study of Bayesian Neural Network Surrogates for Bayesian Optimization
Bayesian optimization is a highly efficient approach to optimizing objective functions which are expensive to query. These objectives are typically represented by Gaussian process (GP) surrogate models which are easy to optimize and support exact inference. While standard GP surrogates have been well-established in Bayesian optimization, Bayesian neural networks (BNNs) have recently become practical function approximators, with many benefits over standard GPs such as the ability to naturally handle non-stationarity and learn representations for high-dimensional data. In this paper, we study BNNs as alternatives to standard GP surrogates for optimization. We consider a variety of approximate inference procedures for finite-width BNNs, including high-quality Hamiltonian Monte Carlo, low-cost stochastic MCMC, and heuristics such as deep ensembles. We also consider infinite-width BNNs and partially stochastic models such as deep kernel learning. We evaluate this collection of surrogate models on diverse problems with varying dimensionality, number of objectives, non-stationarity, and discrete and continuous inputs. We find: (i) the ranking of methods is highly problem dependent, suggesting the need for tailored inductive biases; (ii) HMC is the most successful approximate inference procedure for fully stochastic BNNs; (iii) full stochasticity may be unnecessary as deep kernel learning is relatively competitive; (iv) infinite-width BNNs are particularly promising, especially in high dimensions.
☆ Variational $f$-Divergence and Derangements for Discriminative Mutual Information Estimation
The accurate estimation of the mutual information is a crucial task in various applications, including machine learning, communications, and biology, since it enables the understanding of complex systems. High-dimensional data render the task extremely challenging due to the amount of data to be processed and the presence of convoluted patterns. Neural estimators based on variational lower bounds of the mutual information have gained attention in recent years but they are prone to either high bias or high variance as a consequence of the partition function. We propose a novel class of discriminative mutual information estimators based on the variational representation of the $f$-divergence. We investigate the impact of the permutation function used to obtain the marginal training samples and present a novel architectural solution based on derangements. The proposed estimator is flexible as it exhibits an excellent bias/variance trade-off. Experiments on reference scenarios demonstrate that our approach outperforms state-of-the-art neural estimators both in terms of accuracy and complexity.
comment: 27 pages, 16 figures
☆ Bias Mitigation Methods for Binary Classification Decision-Making Systems: Survey and Recommendations
Bias mitigation methods for binary classification decision-making systems have been widely researched due to the ever-growing importance of designing fair machine learning processes that are impartial and do not discriminate against individuals or groups based on protected personal characteristics. In this paper, we present a structured overview of the research landscape for bias mitigation methods, report on their benefits and limitations, and provide recommendations for the development of future bias mitigation methods for binary classification.
comment: 22 pages
☆ Monotonic Location Attention for Length Generalization ICML 2023
We explore different ways to utilize position-based cross-attention in seq2seq networks to enable length generalization in algorithmic tasks. We show that a simple approach of interpolating the original and reversed encoded representations combined with relative attention allows near-perfect length generalization for both forward and reverse lookup tasks or copy tasks that had been generally hard to tackle. We also devise harder diagnostic tasks where the relative distance of the ideal attention position varies with timestep. In such settings, the simple interpolation trick with relative attention is not sufficient. We introduce novel variants of location attention building on top of Dubois et al. (2020) to address the new diagnostic tasks. We also show the benefits of our approaches for length generalization in SCAN (Lake & Baroni, 2018) and CFQ (Keysers et al., 2020). Our code is available on GitHub.
comment: Accepted in ICML 2023
☆ Constrained Causal Bayesian Optimization
We propose constrained causal Bayesian optimization (cCBO), an approach for finding interventions in a known causal graph that optimize a target variable under some constraints. cCBO first reduces the search space by exploiting the graph structure and, if available, an observational dataset; and then solves the restricted optimization problem by modelling target and constraint quantities using Gaussian processes and by sequentially selecting interventions via a constrained expected improvement acquisition function. We propose different surrogate models that enable to integrate observational and interventional data while capturing correlation among effects with increasing levels of sophistication. We evaluate cCBO on artificial and real-world causal graphs showing successful trade off between fast convergence and percentage of feasible interventions.
☆ Protein Design with Guided Discrete Diffusion
A popular approach to protein design is to combine a generative model with a discriminative model for conditional sampling. The generative model samples plausible sequences while the discriminative model guides a search for sequences with high fitness. Given its broad success in conditional sampling, classifier-guided diffusion modeling is a promising foundation for protein design, leading many to develop guided diffusion models for structure with inverse folding to recover sequences. In this work, we propose diffusioN Optimized Sampling (NOS), a guidance method for discrete diffusion models that follows gradients in the hidden states of the denoising network. NOS makes it possible to perform design directly in sequence space, circumventing significant limitations of structure-based methods, including scarce data and challenging inverse design. Moreover, we use NOS to generalize LaMBO, a Bayesian optimization procedure for sequence design that facilitates multiple objectives and edit-based constraints. The resulting method, LaMBO-2, enables discrete diffusions and stronger performance with limited edits through a novel application of saliency maps. We apply LaMBO-2 to a real-world protein design task, optimizing antibodies for higher expression yield and binding affinity to a therapeutic target under locality and liability constraints, with 97% expression rate and 25% binding rate in exploratory in vitro experiments.
☆ Learning to solve Bayesian inverse problems: An amortized variational inference approach
Inverse problems, i.e., estimating parameters of physical models from experimental data, are ubiquitous in science and engineering. The Bayesian formulation is the gold standard because it alleviates ill-posedness issues and quantifies epistemic uncertainty. Since analytical posteriors are not typically available, one resorts to Markov chain Monte Carlo sampling or approximate variational inference. However, inference needs to be rerun from scratch for each new set of data. This drawback limits the applicability of the Bayesian formulation to real-time settings, e.g., health monitoring of engineered systems, and medical diagnosis. The objective of this paper is to develop a methodology that enables real-time inference by learning the Bayesian inverse map, i.e., the map from data to posteriors. Our approach is as follows. We represent the posterior distribution using a parameterization based on deep neural networks. Next, we learn the network parameters by amortized variational inference method which involves maximizing the expectation of evidence lower bound over all possible datasets compatible with the model. We demonstrate our approach by solving examples a set of benchmark problems from science and engineering. Our results show that the posterior estimates of our approach are in agreement with the corresponding ground truth obtained by Markov chain Monte Carlo. Once trained, our approach provides the posterior parameters of observation just at the cost of a forward pass of the neural network.
☆ A Novel Black Box Process Quality Optimization Approach based on Hit Rate
Hit rate is a key performance metric in predicting process product quality in integrated industrial processes. It represents the percentage of products accepted by downstream processes within a controlled range of quality. However, optimizing hit rate is a non-convex and challenging problem. To address this issue, we propose a data-driven quasi-convex approach that combines factorial hidden Markov models, multitask elastic net, and quasi-convex optimization. Our approach converts the original non-convex problem into a set of convex feasible problems, achieving an optimal hit rate. We verify the convex optimization property and quasi-convex frontier through Monte Carlo simulations and real-world experiments in steel production. Results demonstrate that our approach outperforms classical models, improving hit rates by at least 41.11% and 31.01% on two real datasets. Furthermore, the quasi-convex frontier provides a reference explanation and visualization for the deterioration of solutions obtained by conventional models.
☆ Representer Point Selection for Explaining Regularized High-dimensional Models ICML 2023
We introduce a novel class of sample-based explanations we term high-dimensional representers, that can be used to explain the predictions of a regularized high-dimensional model in terms of importance weights for each of the training samples. Our workhorse is a novel representer theorem for general regularized high-dimensional models, which decomposes the model prediction in terms of contributions from each of the training samples: with positive (negative) values corresponding to positive (negative) impact training samples to the model's prediction. We derive consequences for the canonical instances of $\ell_1$ regularized sparse models, and nuclear norm regularized low-rank models. As a case study, we further investigate the application of low-rank models in the context of collaborative filtering, where we instantiate high-dimensional representers for specific popular classes of models. Finally, we study the empirical performance of our proposed methods on three real-world binary classification datasets and two recommender system datasets. We also showcase the utility of high-dimensional representers in explaining model recommendations.
comment: Accepted by ICML 2023
☆ Beam Tree Recursive Cells ICML 2023
We propose Beam Tree Recursive Cell (BT-Cell) - a backpropagation-friendly framework to extend Recursive Neural Networks (RvNNs) with beam search for latent structure induction. We further extend this framework by proposing a relaxation of the hard top-k operators in beam search for better propagation of gradient signals. We evaluate our proposed models in different out-of-distribution splits in both synthetic and realistic data. Our experiments show that BTCell achieves near-perfect performance on several challenging structure-sensitive synthetic tasks like ListOps and logical inference while maintaining comparable performance in realistic data against other RvNN-based models. Additionally, we identify a previously unknown failure case for neural models in generalization to unseen number of arguments in ListOps. The code is available at: https://github.com/JRC1995/ BeamTreeRecursiveCells.
comment: Accepted in ICML 2023
☆ Efficient Shapley Values Estimation by Amortization for Text Classification ACL 2023
Despite the popularity of Shapley Values in explaining neural text classification models, computing them is prohibitive for large pretrained models due to a large number of model evaluations. In practice, Shapley Values are often estimated with a small number of stochastic model evaluations. However, we show that the estimated Shapley Values are sensitive to random seed choices -- the top-ranked features often have little overlap across different seeds, especially on examples with longer input texts. This can only be mitigated by aggregating thousands of model evaluations, which on the other hand, induces substantial computational overheads. To mitigate the trade-off between stability and efficiency, we develop an amortized model that directly predicts each input feature's Shapley Value without additional model evaluations. It is trained on a set of examples whose Shapley Values are estimated from a large number of model evaluations to ensure stability. Experimental results on two text classification datasets demonstrate that our amortized model estimates Shapley Values accurately with up to 60 times speedup compared to traditional methods. Furthermore, the estimated values are stable as the inference is deterministic. We release our code at https://github.com/yangalan123/Amortized-Interpretability.
comment: ACL 2023 Camera Ready
☆ A Nested Matrix-Tensor Model for Noisy Multi-view Clustering
In this paper, we propose a nested matrix-tensor model which extends the spiked rank-one tensor model of order three. This model is particularly motivated by a multi-view clustering problem in which multiple noisy observations of each data point are acquired, with potentially non-uniform variances along the views. In this case, data can be naturally represented by an order-three tensor where the views are stacked. Given such a tensor, we consider the estimation of the hidden clusters via performing a best rank-one tensor approximation. In order to study the theoretical performance of this approach, we characterize the behavior of this best rank-one approximation in terms of the alignments of the obtained component vectors with the hidden model parameter vectors, in the large-dimensional regime. In particular, we show that our theoretical results allow us to anticipate the exact accuracy of the proposed clustering approach. Furthermore, numerical experiments indicate that leveraging our tensor-based approach yields better accuracy compared to a naive unfolding-based algorithm which ignores the underlying low-rank tensor structure. Our analysis unveils unexpected and non-trivial phase transition phenomena depending on the model parameters, ``interpolating'' between the typical behavior observed for the spiked matrix and tensor models.
☆ InGram: Inductive Knowledge Graph Embedding via Relation Graphs ICML 2023
Inductive knowledge graph completion has been considered as the task of predicting missing triplets between new entities that are not observed during training. While most inductive knowledge graph completion methods assume that all entities can be new, they do not allow new relations to appear at inference time. This restriction prohibits the existing methods from appropriately handling real-world knowledge graphs where new entities accompany new relations. In this paper, we propose an INductive knowledge GRAph eMbedding method, InGram, that can generate embeddings of new relations as well as new entities at inference time. Given a knowledge graph, we define a relation graph as a weighted graph consisting of relations and the affinity weights between them. Based on the relation graph and the original knowledge graph, InGram learns how to aggregate neighboring embeddings to generate relation and entity embeddings using an attention mechanism. Experimental results show that InGram outperforms 14 different state-of-the-art methods on varied inductive learning scenarios.
comment: 14 pages, 4 figures, 6 tables, 40th International Conference on Machine Learning (ICML 2023)
☆ Adam Accumulation to Reduce Memory Footprints of both Activations and Gradients for Large-scale DNN Training
Running out of GPU memory has become a main bottleneck for large-scale DNN training. How to reduce the memory footprint during training has received intensive research attention. We find that previous gradient accumulation reduces activation memory but fails to be compatible with gradient memory reduction due to a contradiction between preserving gradients and releasing gradients. To address this issue, we propose a novel optimizer accumulation method for Adam, named Adam Accumulation (AdamA), which enables reducing both activation and gradient memory. Specifically, AdamA directly integrates gradients into optimizer states and accumulates optimizer states over micro-batches, so that gradients can be released immediately after use. We mathematically and experimentally demonstrate AdamA yields the same convergence properties as Adam. Evaluated on transformer-based models, AdamA achieves up to 23% memory reduction compared to gradient accumulation with less than 2% degradation in training throughput. Notably, AdamA can work together with memory reduction methods for optimizer states to fit 1.26x~3.14x larger models over PyTorch and DeepSpeed baseline on GPUs with different memory capacities.
☆ Knowledge Graph Embeddings in the Biomedical Domain: Are They Useful? A Look at Link Prediction, Rule Learning, and Downstream Polypharmacy Tasks
Knowledge graphs are powerful tools for representing and organising complex biomedical data. Several knowledge graph embedding algorithms have been proposed to learn from and complete knowledge graphs. However, a recent study demonstrates the limited efficacy of these embedding algorithms when applied to biomedical knowledge graphs, raising the question of whether knowledge graph embeddings have limitations in biomedical settings. This study aims to apply state-of-the-art knowledge graph embedding models in the context of a recent biomedical knowledge graph, BioKG, and evaluate their performance and potential downstream uses. We achieve a three-fold improvement in terms of performance based on the HITS@10 score over previous work on the same biomedical knowledge graph. Additionally, we provide interpretable predictions through a rule-based method. We demonstrate that knowledge graph embedding models are applicable in practice by evaluating the best-performing model on four tasks that represent real-life polypharmacy situations. Results suggest that knowledge learnt from large biomedical knowledge graphs can be transferred to such downstream use cases. Our code is available at https://github.com/aryopg/biokge.
☆ Federated Learning in the Presence of Adversarial Client Unavailability
Federated learning is a decentralized machine learning framework wherein not all clients are able to participate in each round. An emerging line of research is devoted to tackling arbitrary client unavailability. Existing theoretical analysis imposes restrictive structural assumptions on the unavailability patterns, and their proposed algorithms were tailored to those assumptions. In this paper, we relax those assumptions and consider adversarial client unavailability. To quantify the degrees of client unavailability, we use the notion of {\em $\epsilon$-adversary dropout fraction}. For both non-convex and strongly-convex global objectives, we show that simple variants of FedAvg or FedProx, albeit completely agnostic to $\epsilon$, converge to an estimation error on the order of $\epsilon (G^2 + \sigma^2)$, where $G$ is a heterogeneity parameter and $\sigma^2$ is the noise level. We prove that this estimation error is minimax-optimal. We also show that the variants of FedAvg or FedProx have convergence speeds $O(1/\sqrt{T})$ for non-convex objectives and $O(1/T)$ for strongly-convex objectives, both of which are the best possible for any first-order method that only has access to noisy gradients. Our proofs build upon a tight analysis of the selection bias that persists in the entire learning process. We validate our theoretical prediction through numerical experiments on synthetic and real-world datasets.
☆ Analysing high resolution digital Mars images using machine learning
The search for ephemeral liquid water on Mars is an ongoing activity. After the recession of the seasonal polar ice cap on Mars, small water ice patches may be left behind in shady places due to the low thermal conductivity of the Martian surface and atmosphere. During late spring and early summer, these patches may be exposed to direct sunlight and warm up rapidly enough for the liquid phase to emerge. To see the spatial and temporal occurrence of such ice patches, optical images should be searched for and checked. Previously a manual image analysis was conducted on 110 images from the southern hemisphere, captured by the High Resolution Imaging Science Experiment (HiRISE) camera onboard the Mars Reconnaissance Orbiter space mission. Out of these, 37 images were identified with smaller ice patches, which were distinguishable by their brightness, colour and strong connection to local topographic shading. In this study, a convolutional neural network (CNN) is applied to find further images with potential water ice patches in the latitude band between -40{\deg} and -60{\deg}, where the seasonal retreat of the polar ice cap happens. Previously analysed HiRISE images are used to train the model, each was split into hundreds of pieces, expanding the training dataset to 6240 images. A test run conducted on 38 new HiRISE images indicates that the program can generally recognise small bright patches, however further training might be needed for more precise predictions.Using a CNN model may make it realistic to analyse all available surface images, aiding us in selecting areas for further investigation.
☆ MicroSegNet: A Deep Learning Approach for Prostate Segmentation on Micro-Ultrasound Images
Micro-ultrasound (micro-US) is a novel 29-MHz ultrasound technique that provides 3-4 times higher resolution than traditional ultrasound, delivering comparable accuracy for diagnosing prostate cancer to MRI but at a lower cost. Accurate prostate segmentation is crucial for prostate volume measurement, cancer diagnosis, prostate biopsy, and treatment planning. This paper proposes a deep learning approach for automated, fast, and accurate prostate segmentation on micro-US images. Prostate segmentation on micro-US is challenging due to artifacts and indistinct borders between the prostate, bladder, and urethra in the midline. We introduce MicroSegNet, a multi-scale annotation-guided Transformer UNet model to address this challenge. During the training process, MicroSegNet focuses more on regions that are hard to segment (challenging regions), where expert and non-expert annotations show discrepancies. We achieve this by proposing an annotation-guided cross entropy loss that assigns larger weight to pixels in hard regions and lower weight to pixels in easy regions. We trained our model using micro-US images from 55 patients, followed by evaluation on 20 patients. Our MicroSegNet model achieved a Dice coefficient of 0.942 and a Hausdorff distance of 2.11 mm, outperforming several state-of-the-art segmentation methods, as well as three human annotators with different experience levels. We will make our code and dataset publicly available to promote transparency and collaboration in research.
☆ Multi-Dataset Co-Training with Sharpness-Aware Optimization for Audio Anti-spoofing
Audio anti-spoofing for automatic speaker verification aims to safeguard users' identities from spoofing attacks. Although state-of-the-art spoofing countermeasure(CM) models perform well on specific datasets, they lack generalization when evaluated with different datasets. To address this limitation, previous studies have explored large pre-trained models, which require significant resources and time. We aim to develop a compact but well-generalizing CM model that can compete with large pre-trained models. Our approach involves multi-dataset co-training and sharpness-aware minimization, which has not been investigated in this domain. Extensive experiments reveal that proposed method yield competitive results across various datasets while utilizing 4,000 times less parameters than the large pre-trained models.
comment: Interspeech 2023
☆ Not All Neuro-Symbolic Concepts Are Created Equal: Analysis and Mitigation of Reasoning Shortcuts
Neuro-Symbolic (NeSy) predictive models hold the promise of improved compliance with given constraints, systematic generalization, and interpretability, as they allow to infer labels that are consistent with some prior knowledge by reasoning over high-level concepts extracted from sub-symbolic inputs. It was recently shown that NeSy predictors are affected by reasoning shortcuts: they can attain high accuracy but by leveraging concepts with unintended semantics, thus coming short of their promised advantages. Yet, a systematic characterization of reasoning shortcuts and of potential mitigation strategies is missing. This work fills this gap by characterizing them as unintended optima of the learning objective and identifying four key conditions behind their occurrence. Based on this, we derive several natural mitigation strategies, and analyze their efficacy both theoretically and empirically. Our analysis shows reasoning shortcuts are difficult to deal with, casting doubts on the trustworthiness and interpretability of existing NeSy solutions.
☆ A Geometric Perspective on Diffusion Models
Recent years have witnessed significant progress in developing efficient training and fast sampling approaches for diffusion models. A recent remarkable advancement is the use of stochastic differential equations (SDEs) to describe data perturbation and generative modeling in a unified mathematical framework. In this paper, we reveal several intriguing geometric structures of diffusion models and contribute a simple yet powerful interpretation to their sampling dynamics. Through carefully inspecting a popular variance-exploding SDE and its marginal-preserving ordinary differential equation (ODE) for sampling, we discover that the data distribution and the noise distribution are smoothly connected with an explicit, quasi-linear sampling trajectory, and another implicit denoising trajectory, which even converges faster in terms of visual quality. We also establish a theoretical relationship between the optimal ODE-based sampling and the classic mean-shift (mode-seeking) algorithm, with which we can characterize the asymptotic behavior of diffusion models and identify the score deviation. These new geometric observations enable us to improve previous sampling algorithms, re-examine latent interpolation, as well as re-explain the working principles of distillation-based fast sampling techniques.
☆ Image Registration of In Vivo Micro-Ultrasound and Ex Vivo Pseudo-Whole Mount Histopathology Images of the Prostate: A Proof-of-Concept Study
Early diagnosis of prostate cancer significantly improves a patient's 5-year survival rate. Biopsy of small prostate cancers is improved with image-guided biopsy. MRI-ultrasound fusion-guided biopsy is sensitive to smaller tumors but is underutilized due to the high cost of MRI and fusion equipment. Micro-ultrasound (micro-US), a novel high-resolution ultrasound technology, provides a cost-effective alternative to MRI while delivering comparable diagnostic accuracy. However, the interpretation of micro-US is challenging due to subtle gray scale changes indicating cancer vs normal tissue. This challenge can be addressed by training urologists with a large dataset of micro-US images containing the ground truth cancer outlines. Such a dataset can be mapped from surgical specimens (histopathology) onto micro-US images via image registration. In this paper, we present a semi-automated pipeline for registering in vivo micro-US images with ex vivo whole-mount histopathology images. Our pipeline begins with the reconstruction of pseudo-whole-mount histopathology images and a 3D micro-US volume. Each pseudo-whole-mount histopathology image is then registered with the corresponding axial micro-US slice using a two-stage approach that estimates an affine transformation followed by a deformable transformation. We evaluated our registration pipeline using micro-US and histopathology images from 18 patients who underwent radical prostatectomy. The results showed a Dice coefficient of 0.94 and a landmark error of 2.7 mm, indicating the accuracy of our registration pipeline. This proof-of-concept study demonstrates the feasibility of accurately aligning micro-US and histopathology images. To promote transparency and collaboration in research, we will make our code and dataset publicly available.
☆ MetaDiffuser: Diffusion Model as Conditional Planner for Offline Meta-RL ICML 23
Recently, diffusion model shines as a promising backbone for the sequence modeling paradigm in offline reinforcement learning(RL). However, these works mostly lack the generalization ability across tasks with reward or dynamics change. To tackle this challenge, in this paper we propose a task-oriented conditioned diffusion planner for offline meta-RL(MetaDiffuser), which considers the generalization problem as conditional trajectory generation task with contextual representation. The key is to learn a context conditioned diffusion model which can generate task-oriented trajectories for planning across diverse tasks. To enhance the dynamics consistency of the generated trajectories while encouraging trajectories to achieve high returns, we further design a dual-guided module in the sampling process of the diffusion model. The proposed framework enjoys the robustness to the quality of collected warm-start data from the testing task and the flexibility to incorporate with different task representation method. The experiment results on MuJoCo benchmarks show that MetaDiffuser outperforms other strong offline meta-RL baselines, demonstrating the outstanding conditional generation ability of diffusion architecture.
comment: 19 pages, 4 figures, accepted by ICML 23'
☆ Representation-Driven Reinforcement Learning ICML 2023
We present a representation-driven framework for reinforcement learning. By representing policies as estimates of their expected values, we leverage techniques from contextual bandits to guide exploration and exploitation. Particularly, embedding a policy network into a linear feature space allows us to reframe the exploration-exploitation problem as a representation-exploitation problem, where good policy representations enable optimal exploration. We demonstrate the effectiveness of this framework through its application to evolutionary and policy gradient-based approaches, leading to significantly improved performance compared to traditional methods. Our framework provides a new perspective on reinforcement learning, highlighting the importance of policy representation in determining optimal exploration-exploitation strategies.
comment: Accepted to ICML 2023
☆ Fully Dynamic Submodular Maximization over Matroids ICML 2023
Maximizing monotone submodular functions under a matroid constraint is a classic algorithmic problem with multiple applications in data mining and machine learning. We study this classic problem in the fully dynamic setting, where elements can be both inserted and deleted in real-time. Our main result is a randomized algorithm that maintains an efficient data structure with an $\tilde{O}(k^2)$ amortized update time (in the number of additions and deletions) and yields a $4$-approximate solution, where $k$ is the rank of the matroid.
comment: Accepted at ICML 2023
☆ Are Neural Operators Really Neural Operators? Frame Theory Meets Operator Learning
Recently, there has been significant interest in operator learning, i.e. learning mappings between infinite-dimensional function spaces. This has been particularly relevant in the context of learning partial differential equations from data. However, it has been observed that proposed models may not behave as operators when implemented on a computer, questioning the very essence of what operator learning should be. We contend that in addition to defining the operator at the continuous level, some form of continuous-discrete equivalence is necessary for an architecture to genuinely learn the underlying operator, rather than just discretizations of it. To this end, we propose to employ frames, a concept in applied harmonic analysis and signal processing that gives rise to exact and stable discrete representations of continuous signals. Extending these concepts to operators, we introduce a unifying mathematical framework of Representation equivalent Neural Operator (ReNO) to ensure operations at the continuous and discrete level are equivalent. Lack of this equivalence is quantified in terms of aliasing errors. We analyze various existing operator learning architectures to determine whether they fall within this framework, and highlight implications when they fail to do so.
comment: 21 pages
☆ Neuron to Graph: Interpreting Language Model Neurons at Scale
Advances in Large Language Models (LLMs) have led to remarkable capabilities, yet their inner mechanisms remain largely unknown. To understand these models, we need to unravel the functions of individual neurons and their contribution to the network. This paper introduces a novel automated approach designed to scale interpretability techniques across a vast array of neurons within LLMs, to make them more interpretable and ultimately safe. Conventional methods require examination of examples with strong neuron activation and manual identification of patterns to decipher the concepts a neuron responds to. We propose Neuron to Graph (N2G), an innovative tool that automatically extracts a neuron's behaviour from the dataset it was trained on and translates it into an interpretable graph. N2G uses truncation and saliency methods to emphasise only the most pertinent tokens to a neuron while enriching dataset examples with diverse samples to better encompass the full spectrum of neuron behaviour. These graphs can be visualised to aid researchers' manual interpretation, and can generate token activations on text for automatic validation by comparison with the neuron's ground truth activations, which we use to show that the model is better at predicting neuron activation than two baseline methods. We also demonstrate how the generated graph representations can be flexibly used to facilitate further automation of interpretability research, by searching for neurons with particular properties, or programmatically comparing neurons to each other to identify similar neurons. Our method easily scales to build graph representations for all neurons in a 6-layer Transformer model using a single Tesla T4 GPU, allowing for wide usability. We release the code and instructions for use at https://github.com/alexjfoote/Neuron2Graph.
☆ Improving Expressivity of GNNs with Subgraph-specific Factor Embedded Normalization
Graph Neural Networks~(GNNs) have emerged as a powerful category of learning architecture for handling graph-structured data. However, existing GNNs typically ignore crucial structural characteristics in node-induced subgraphs, which thus limits their expressiveness for various downstream tasks. In this paper, we strive to strengthen the representative capabilities of GNNs by devising a dedicated plug-and-play normalization scheme, termed as~\emph{\textbf{SU}bgraph-s\textbf{PE}cific Facto\textbf{R} Embedded Normalization}~(SuperNorm), that explicitly considers the intra-connection information within each node-induced subgraph. To this end, we embed the subgraph-specific factor at the beginning and the end of the standard BatchNorm, as well as incorporate graph instance-specific statistics for improved distinguishable capabilities. In the meantime, we provide theoretical analysis to support that, with the elaborated SuperNorm, an arbitrary GNN is at least as powerful as the 1-WL test in distinguishing non-isomorphism graphs. Furthermore, the proposed SuperNorm scheme is also demonstrated to alleviate the over-smoothing phenomenon. Experimental results related to predictions of graph, node, and link properties on the eight popular datasets demonstrate the effectiveness of the proposed method. The code is available at \url{https://github.com/chenchkx/SuperNorm}.
comment: 13 pages, 7 figures
☆ Adaptive Conformal Regression with Jackknife+ Rescaled Scores
Conformal regression provides prediction intervals with global coverage guarantees, but often fails to capture local error distributions, leading to non-homogeneous coverage. We address this with a new adaptive method based on rescaling conformal scores with an estimate of local score distribution, inspired by the Jackknife+ method, which enables the use of calibration data in conformal scores without breaking calibration-test exchangeability. Our approach ensures formal global coverage guarantees and is supported by new theoretical results on local coverage, including an a posteriori bound on any calibration score. The strength of our approach lies in achieving local coverage without sacrificing calibration set size, improving the applicability of conformal prediction intervals in various settings. As a result, our method provides prediction intervals that outperform previous methods, particularly in the low-data regime, making it especially relevant for real-world applications such as healthcare and biomedical domains where uncertainty needs to be quantified accurately despite low sample data.
comment: 24 pages, 7 figures
☆ fpgaHART: A toolflow for throughput-oriented acceleration of 3D CNNs for HAR onto FPGAs
Surveillance systems, autonomous vehicles, human monitoring systems, and video retrieval are just few of the many applications in which 3D Convolutional Neural Networks are exploited. However, their extensive use is restricted by their high computational and memory requirements, especially when integrated into systems with limited resources. This study proposes a toolflow that optimises the mapping of 3D CNN models for Human Action Recognition onto FPGA devices, taking into account FPGA resources and off-chip memory characteristics. The proposed system employs Synchronous Dataflow (SDF) graphs to model the designs and introduces transformations to expand and explore the design space, resulting in high-throughput designs. A variety of 3D CNN models were evaluated using the proposed toolflow on multiple FPGA devices, demonstrating its potential to deliver competitive performance compared to earlier hand-tuned and model-specific designs.
comment: 7 pages, 3 figures, 4 tables. arXiv admin note: substantial text overlap with arXiv:2305.18479
☆ Handling Large Discrete Action Spaces via Dynamic Neighborhood Construction
Large discrete action spaces remain a central challenge for reinforcement learning methods. Such spaces are encountered in many real-world applications, e.g., recommender systems, multi-step planning, and inventory replenishment. The mapping of continuous proxies to discrete actions is a promising paradigm for handling large discrete action spaces. Existing continuous-to-discrete mapping approaches involve searching for discrete neighboring actions in a static pre-defined neighborhood, which requires discrete neighbor lookups across the entire action space. Hence, scalability issues persist. To mitigate this drawback, we propose a novel Dynamic Neighborhood Construction (DNC) method, which dynamically constructs a discrete neighborhood to map the continuous proxy, thus efficiently exploiting the underlying action space. We demonstrate the robustness of our method by benchmarking it against three state-of-the-art approaches designed for large discrete action spaces across three different environments. Our results show that DNC matches or outperforms state-of-the-art approaches while being more computationally efficient. Furthermore, our method scales to action spaces that so far remained computationally intractable for existing methodologies.
☆ Evaluating Machine Learning Models with NERO: Non-Equivariance Revealed on Orbits
Proper evaluations are crucial for better understanding, troubleshooting, interpreting model behaviors and further improving model performance. While using scalar-based error metrics provides a fast way to overview model performance, they are often too abstract to display certain weak spots and lack information regarding important model properties, such as robustness. This not only hinders machine learning models from being more interpretable and gaining trust, but also can be misleading to both model developers and users. Additionally, conventional evaluation procedures often leave researchers unclear about where and how model fails, which complicates model comparisons and further developments. To address these issues, we propose a novel evaluation workflow, named Non-Equivariance Revealed on Orbits (NERO) Evaluation. The goal of NERO evaluation is to turn focus from traditional scalar-based metrics onto evaluating and visualizing models equivariance, closely capturing model robustness, as well as to allow researchers quickly investigating interesting or unexpected model behaviors. NERO evaluation is consist of a task-agnostic interactive interface and a set of visualizations, called NERO plots, which reveals the equivariance property of the model. Case studies on how NERO evaluation can be applied to multiple research areas, including 2D digit recognition, object detection, particle image velocimetry (PIV), and 3D point cloud classification, demonstrate that NERO evaluation can quickly illustrate different model equivariance, and effectively explain model behaviors through interactive visualizations of the model outputs. In addition, we propose consensus, an alternative to ground truths, to be used in NERO evaluation so that model equivariance can still be evaluated with new, unlabeled datasets.
☆ Spectral Heterogeneous Graph Convolutions via Positive Noncommutative Polynomials
Heterogeneous Graph Neural Networks (HGNNs) have gained significant popularity in various heterogeneous graph learning tasks. However, most HGNNs rely on spatial domain-based message passing and attention modules for information propagation and aggregation. These spatial-based HGNNs neglect the utilization of spectral graph convolutions, which are the foundation of Graph Convolutional Networks (GCN) on homogeneous graphs. Inspired by the effectiveness and scalability of spectral-based GNNs on homogeneous graphs, this paper explores the extension of spectral-based GNNs to heterogeneous graphs. We propose PSHGCN, a novel heterogeneous convolutional network based on positive noncommutative polynomials. PSHGCN provides a simple yet effective approach for learning spectral graph convolutions on heterogeneous graphs. Moreover, we demonstrate the rationale of PSHGCN in graph optimization. We conducted an extensive experimental study to show that PSHGCN can learn diverse spectral heterogeneous graph convolutions and achieve superior performance in node classification tasks. Our code is available at https://github.com/ivam-he/PSHGCN.
comment: 10 pages
☆ There is more to graphs than meets the eye: Learning universal features with self-supervision
We study the problem of learning universal features across multiple graphs through self-supervision. Graph self supervised learning has been shown to facilitate representation learning, and produce competitive models compared to supervised baselines. However, existing methods of self-supervision learn features from one graph, and thus, produce models that are specialized to a particular graph. We hypothesize that leveraging multiple graphs of the same type/class can improve the quality of learnt representations in the model by extracting features that are universal to the class of graphs. We adopt a transformer backbone that acts as a universal representation learning module for multiple graphs. We leverage neighborhood aggregation coupled with graph-specific embedding generator to transform disparate node embeddings from multiple graphs to a common space for the universal backbone. We learn both universal and graph-specific parameters in an end-to-end manner. Our experiments reveal that leveraging multiple graphs of the same type -- citation networks -- improves the quality of representations and results in better performance on downstream node classification task compared to self-supervision with one graph. The results of our study improve the state-of-the-art in graph self-supervised learning, and bridge the gap between self-supervised and supervised performance.
comment: arXiv admin note: text overlap with arXiv:2302.11939, arXiv:2301.13287, arXiv:2305.12686, arXiv:2305.02299
☆ Designing Closed-Loop Models for Task Allocation
Automatically assigning tasks to people is challenging because human performance can vary across tasks for many reasons. This challenge is further compounded in real-life settings in which no oracle exists to assess the quality of human decisions and task assignments made. Instead, we find ourselves in a "closed" decision-making loop in which the same fallible human decisions we rely on in practice must also be used to guide task allocation. How can imperfect and potentially biased human decisions train an accurate allocation model? Our key insight is to exploit weak prior information on human-task similarity to bootstrap model training. We show that the use of such a weak prior can improve task allocation accuracy, even when human decision-makers are fallible and biased. We present both theoretical analysis and empirical evaluation over synthetic data and a social media toxicity detection task. Results demonstrate the efficacy of our approach.
comment: Accepted for publication in the International Conference on Hybrid Human-Artificial Intelligence (HHAI) 2023
☆ Relaxing the Additivity Constraints in Decentralized No-Regret High-Dimensional Bayesian Optimization
Bayesian Optimization (BO) is typically used to optimize an unknown function $f$ that is noisy and costly to evaluate, by exploiting an acquisition function that must be maximized at each optimization step. Although provably asymptotically optimal BO algorithms are efficient at optimizing low-dimensional functions, scaling them to high-dimensional spaces remains an open research problem, often tackled by assuming an additive structure for $f$. However, such algorithms introduce additional restrictive assumptions on the additive structure that reduce their applicability domain. In this paper, we relax the restrictive assumptions on the additive structure of $f$, at the expense of weakening the maximization guarantees of the acquisition function, and we address the over-exploration problem for decentralized BO algorithms. To these ends, we propose DuMBO, an asymptotically optimal decentralized BO algorithm that achieves very competitive performance against state-of-the-art BO algorithms, especially when the additive structure of $f$ does not exist or comprises high-dimensional factors.
☆ EAMDrift: An interpretable self retrain model for time series ECML
The use of machine learning for time series prediction has become increasingly popular across various industries thanks to the availability of time series data and advancements in machine learning algorithms. However, traditional methods for time series forecasting rely on pre-optimized models that are ill-equipped to handle unpredictable patterns in data. In this paper, we present EAMDrift, a novel method that combines forecasts from multiple individual predictors by weighting each prediction according to a performance metric. EAMDrift is designed to automatically adapt to out-of-distribution patterns in data and identify the most appropriate models to use at each moment through interpretable mechanisms, which include an automatic retraining process. Specifically, we encode different concepts with different models, each functioning as an observer of specific behaviors. The activation of the overall model then identifies which subset of the concept observers is identifying concepts in the data. This activation is interpretable and based on learned rules, allowing to study of input variables relations. Our study on real-world datasets shows that EAMDrift outperforms individual baseline models by 20% and achieves comparable accuracy results to non-interpretable ensemble models. These findings demonstrate the efficacy of EAMDrift for time-series prediction and highlight the importance of interpretability in machine learning models.
comment: Submitted to ECML PKDD 2023
☆ An Empirical Study of Federated Learning on IoT-Edge Devices: Resource Allocation and Heterogeneity
Nowadays, billions of phones, IoT and edge devices around the world generate data continuously, enabling many Machine Learning (ML)-based products and applications. However, due to increasing privacy concerns and regulations, these data tend to reside on devices (clients) instead of being centralized for performing traditional ML model training. Federated Learning (FL) is a distributed approach in which a single server and multiple clients collaboratively build an ML model without moving data away from clients. Whereas existing studies on FL have their own experimental evaluations, most experiments were conducted using a simulation setting or a small-scale testbed. This might limit the understanding of FL implementation in realistic environments. In this empirical study, we systematically conduct extensive experiments on a large network of IoT and edge devices (called IoT-Edge devices) to present FL real-world characteristics, including learning performance and operation (computation and communication) costs. Moreover, we mainly concentrate on heterogeneous scenarios, which is the most challenging issue of FL. By investigating the feasibility of on-device implementation, our study provides valuable insights for researchers and practitioners, promoting the practicality of FL and assisting in improving the current design of real FL systems.
☆ Bridging Spectral Embedding and Matrix Completion in Self-Supervised Learning
Self-supervised methods received tremendous attention thanks to their seemingly heuristic approach to learning representations that respect the semantics of the data without any apparent supervision in the form of labels. A growing body of literature is already being published in an attempt to build a coherent and theoretically grounded understanding of the workings of a zoo of losses used in modern self-supervised representation learning methods. In this paper, we attempt to provide an understanding from the perspective of a Laplace operator and connect the inductive bias stemming from the augmentation process to a low-rank matrix completion problem. To this end, we leverage the results from low-rank matrix completion to provide theoretical analysis on the convergence of modern SSL methods and a key property that affects their downstream performance.
comment: 13 pages, 3 figures
☆ Direct Diffusion Bridge using Data Consistency for Inverse Problems
Diffusion model-based inverse problem solvers have shown impressive performance, but are limited in speed, mostly as they require reverse diffusion sampling starting from noise. Several recent works have tried to alleviate this problem by building a diffusion process, directly bridging the clean and the corrupted for specific inverse problems. In this paper, we first unify these existing works under the name Direct Diffusion Bridges (DDB), showing that while motivated by different theories, the resulting algorithms only differ in the choice of parameters. Then, we highlight a critical limitation of the current DDB framework, namely that it does not ensure data consistency. To address this problem, we propose a modified inference procedure that imposes data consistency without the need for fine-tuning. We term the resulting method data Consistent DDB (CDDB), which outperforms its inconsistent counterpart in terms of both perception and distortion metrics, thereby effectively pushing the Pareto-frontier toward the optimum. Our proposed method achieves state-of-the-art results on both evaluation criteria, showcasing its superiority over existing methods.
comment: 16 pages, 6 figures
☆ Distance Rank Score: Unsupervised filter method for feature selection on imbalanced dataset
This paper presents a new filter method for unsupervised feature selection. This method is particularly effective on imbalanced multi-class dataset, as in case of clusters of different anomaly types. Existing methods usually involve the variance of the features, which is not suitable when the different types of observations are not represented equally. Our method, based on Spearman's Rank Correlation between distances on the observations and on feature values, avoids this drawback. The performance of the method is measured on several clustering problems and is compared with existing filter methods suitable for unsupervised data.
☆ Neuro-Causal Factor Analysis
Factor analysis (FA) is a statistical tool for studying how observed variables with some mutual dependences can be expressed as functions of mutually independent unobserved factors, and it is widely applied throughout the psychological, biological, and physical sciences. We revisit this classic method from the comparatively new perspective given by advancements in causal discovery and deep learning, introducing a framework for Neuro-Causal Factor Analysis (NCFA). Our approach is fully nonparametric: it identifies factors via latent causal discovery methods and then uses a variational autoencoder (VAE) that is constrained to abide by the Markov factorization of the distribution with respect to the learned graph. We evaluate NCFA on real and synthetic data sets, finding that it performs comparably to standard VAEs on data reconstruction tasks but with the advantages of sparser architecture, lower model complexity, and causal interpretability. Unlike traditional FA methods, our proposed NCFA method allows learning and reasoning about the latent factors underlying observed data from a justifiably causal perspective, even when the relations between factors and measurements are highly nonlinear.
comment: 23 pages, 13 figures
☆ Primal-Attention: Self-attention through Asymmetric Kernel SVD in Primal Representation
Recently, a new line of works has emerged to understand and improve self-attention in Transformers by treating it as a kernel machine. However, existing works apply the methods for symmetric kernels to the asymmetric self-attention, resulting in a nontrivial gap between the analytical understanding and numerical implementation. In this paper, we provide a new perspective to represent and optimize self-attention through asymmetric Kernel Singular Value Decomposition (KSVD), which is also motivated by the low-rank property of self-attention normally observed in deep layers. Through asymmetric KSVD, $i$) a primal-dual representation of self-attention is formulated, where the optimization objective is cast to maximize the projection variances in the attention outputs; $ii$) a novel attention mechanism, i.e., Primal-Attention, is proposed via the primal representation of KSVD, avoiding explicit computation of the kernel matrix in the dual; $iii$) with KKT conditions, we prove that the stationary solution to the KSVD optimization in Primal-Attention yields a zero-value objective. In this manner, KSVD optimization can be implemented by simply minimizing a regularization loss, so that low-rank property is promoted without extra decomposition. Numerical experiments show state-of-the-art performance of our Primal-Attention with improved efficiency. Moreover, we demonstrate that the deployed KSVD optimization regularizes Primal-Attention with a sharper singular value decay than that of the canonical self-attention, further verifying the great potential of our method. To the best of our knowledge, this is the first work that provides a primal-dual representation for the asymmetric kernel in self-attention and successfully applies it to modeling and optimization.
comment: We provide a primal-dual representation for the asymmetric self-attention in transformer that allows to avoid explicit computation of the kernel matrix
☆ Deep learning and MCMC with aggVAE for shifting administrative boundaries: mapping malaria prevalence in Kenya
Model-based disease mapping remains a fundamental policy-informing tool in public health and disease surveillance with hierarchical Bayesian models being the current state-of-the-art approach. When working with areal data, e.g. aggregates at the administrative unit level such as district or province, routinely used models rely on the adjacency structure of areal units to account for spatial correlations. The goal of disease surveillance systems is to track disease outcomes over time, but this provides challenging in situations of crises, such as political changes, leading to changes of administrative boundaries. Kenya is an example of such country. Moreover, adjacency-based approach ignores the continuous nature of spatial processes and cannot solve the change-of-support problem, i.e. when administrative boundaries change. We present a novel, practical, and easy to implement solution relying on a methodology combining deep generative modelling and fully Bayesian inference. We build on the recent work of PriorVAE able to encode spatial priors over small areas with variational autoencoders, to map malaria prevalence in Kenya. We solve the change-of-support problem arising from Kenya changing its district boundaries in 2010. We draw realisations of the Gaussian Process (GP) prior over a fine artificial spatial grid representing continuous space and then aggregate these realisations to the level of administrative boundaries. The aggregated values are then encoded using the PriorVAE technique. The trained priors (aggVAE) are then used at the inference stage instead of the GP priors within a Markov chain Monte Carlo (MCMC) scheme. We demonstrate that it is possible to use the flexible and appropriate model for areal data based on aggregation of continuous priors, and that inference is orders of magnitude faster when using aggVAE than combining the original GP priors and the aggregation step.
☆ Off-By-One Implementation Error in J-UNIWARD
J-UNIWARD is a popular steganography method for hiding secret messages in JPEG cover images. As a content-adaptive method, J-UNIWARD aims to embed into textured image regions where changes are difficult to detect. To this end, J-UNIWARD first assigns to each DCT coefficient an embedding cost calculated based on the image's Wavelet residual, and then uses a coding method that minimizes the cost while embedding the desired payload. Changing one DCT coefficient affects a 23x23 window of Wavelet coefficients. To speed up the costmap computation, the original implementation pre-computes the Wavelet residual and then considers per changed DCT coefficient a 23x23 window of the Wavelet residual. However, the implementation accesses a window accidentally shifted by one pixel to the bottom right. In this report, we evaluate the effect of this off-by-one error on the resulting costmaps. Some image blocks are over-priced while other image blocks are under-priced, but the difference is relatively small. The off-by-one error seems to make little difference for learning-based steganalysis.
☆ Ambiguity in solving imaging inverse problems with deep learning based operators
In recent years, large convolutional neural networks have been widely used as tools for image deblurring, because of their ability in restoring images very precisely. It is well known that image deblurring is mathematically modeled as an ill-posed inverse problem and its solution is difficult to approximate when noise affects the data. Really, one limitation of neural networks for deblurring is their sensitivity to noise and other perturbations, which can lead to instability and produce poor reconstructions. In addition, networks do not necessarily take into account the numerical formulation of the underlying imaging problem, when trained end-to-end. In this paper, we propose some strategies to improve stability without losing to much accuracy to deblur images with deep-learning based methods. First, we suggest a very small neural architecture, which reduces the execution time for training, satisfying a green AI need, and does not extremely amplify noise in the computed image. Second, we introduce a unified framework where a pre-processing step balances the lack of stability of the following, neural network-based, step. Two different pre-processors are presented: the former implements a strong parameter-free denoiser, and the latter is a variational model-based regularized formulation of the latent imaging problem. This framework is also formally characterized by mathematical analysis. Numerical experiments are performed to verify the accuracy and stability of the proposed approaches for image deblurring when unknown or not-quantified noise is present; the results confirm that they improve the network stability with respect to noise. In particular, the model-based framework represents the most reliable trade-off between visual precision and robustness.
☆ Quality In / Quality Out: Assessing Data quality in an Anomaly Detection Benchmark
Autonomous or self-driving networks are expected to provide a solution to the myriad of extremely demanding new applications in the Future Internet. The key to handle complexity is to perform tasks like network optimization and failure recovery with minimal human supervision. For this purpose, the community relies on the development of new Machine Learning (ML) models and techniques. However, ML can only be as good as the data it is fitted with. Datasets provided to the community as benchmarks for research purposes, which have a relevant impact in research findings and directions, are often assumed to be of good quality by default. In this paper, we show that relatively minor modifications on the same benchmark dataset (UGR'16, a flow-based real-traffic dataset for anomaly detection) cause significantly more impact on model performance than the specific ML technique considered. To understand this finding, we contribute a methodology to investigate the root causes for those differences, and to assess the quality of the data labelling. Our findings illustrate the need to devote more attention into (automatic) data quality assessment and optimization techniques in the context of autonomous networks.
☆ Attention-Based Methods For Audio Question Answering
Audio question answering (AQA) is the task of producing natural language answers when a system is provided with audio and natural language questions. In this paper, we propose neural network architectures based on self-attention and cross-attention for the AQA task. The self-attention layers extract powerful audio and textual representations. The cross-attention maps audio features that are relevant to the textual features to produce answers. All our models are trained on the recently proposed Clotho-AQA dataset for both binary yes/no questions and single-word answer questions. Our results clearly show improvement over the reference method reported in the original paper. On the yes/no binary classification task, our proposed model achieves an accuracy of 68.3% compared to 62.7% in the reference model. For the single-word answers multiclass classifier, our model produces a top-1 and top-5 accuracy of 57.9% and 99.8% compared to 54.2% and 93.7% in the reference model respectively. We further discuss some of the challenges in the Clotho-AQA dataset such as the presence of the same answer word in multiple tenses, singular and plural forms, and the presence of specific and generic answers to the same question. We address these issues and present a revised version of the dataset.
☆ A Bayesian Perspective On Training Data Attribution
Training data attribution (TDA) techniques find influential training data for the model's prediction on the test data of interest. They approximate the impact of down- or up-weighting a particular training sample. While conceptually useful, they are hardly applicable in practice, particularly because of their sensitivity to different model initialisation. In this paper, we introduce a Bayesian perspective on the TDA task, where the learned model is treated as a Bayesian posterior and the TDA estimates as random variables. From this novel viewpoint, we observe that the influence of an individual training sample is often overshadowed by the noise stemming from model initialisation and SGD batch composition. Based on this observation, we argue that TDA can only be reliably used for explaining model predictions that are consistently influenced by certain training data, independent of other noise factors. Our experiments demonstrate the rarity of such noise-independent training-test data pairs but confirm their existence. We recommend that future researchers and practitioners trust TDA estimates only in such cases. Further, we find a disagreement between ground truth and estimated TDA distributions and encourage future work to study this gap. Code is provided at https://github.com/ElisaNguyen/bayesian-tda.
☆ Recursive Metropolis-Hastings Naming Game: Symbol Emergence in a Multi-agent System based on Probabilistic Generative Models
In the studies on symbol emergence and emergent communication in a population of agents, a computational model was employed in which agents participate in various language games. Among these, the Metropolis-Hastings naming game (MHNG) possesses a notable mathematical property: symbol emergence through MHNG is proven to be a decentralized Bayesian inference of representations shared by the agents. However, the previously proposed MHNG is limited to a two-agent scenario. This paper extends MHNG to an N-agent scenario. The main contributions of this paper are twofold: (1) we propose the recursive Metropolis-Hastings naming game (RMHNG) as an N-agent version of MHNG and demonstrate that RMHNG is an approximate Bayesian inference method for the posterior distribution over a latent variable shared by agents, similar to MHNG; and (2) we empirically evaluate the performance of RMHNG on synthetic and real image data, enabling multiple agents to develop and share a symbol system. Furthermore, we introduce two types of approximations -- one-sample and limited-length -- to reduce computational complexity while maintaining the ability to explain communication in a population of agents. The experimental findings showcased the efficacy of RMHNG as a decentralized Bayesian inference for approximating the posterior distribution concerning latent variables, which are jointly shared among agents, akin to MHNG. Moreover, the utilization of RMHNG elucidated the agents' capacity to exchange symbols. Furthermore, the study discovered that even the computationally simplified version of RMHNG could enable symbols to emerge among the agents.
☆ The Tunnel Effect: Building Data Representations in Deep Neural Networks
Deep neural networks are widely known for their remarkable effectiveness across various tasks, with the consensus that deeper networks implicitly learn more complex data representations. This paper shows that sufficiently deep networks trained for supervised image classification split into two distinct parts that contribute to the resulting data representations differently. The initial layers create linearly-separable representations, while the subsequent layers, which we refer to as \textit{the tunnel}, compress these representations and have a minimal impact on the overall performance. We explore the tunnel's behavior through comprehensive empirical studies, highlighting that it emerges early in the training process. Its depth depends on the relation between the network's capacity and task complexity. Furthermore, we show that the tunnel degrades out-of-distribution generalization and discuss its implications for continual learning.
☆ Neural Markov Jump Processes
Markov jump processes are continuous-time stochastic processes with a wide range of applications in both natural and social sciences. Despite their widespread use, inference in these models is highly non-trivial and typically proceeds via either Monte Carlo or expectation-maximization methods. In this work we introduce an alternative, variational inference algorithm for Markov jump processes which relies on neural ordinary differential equations, and is trainable via back-propagation. Our methodology learns neural, continuous-time representations of the observed data, that are used to approximate the initial distribution and time-dependent transition probability rates of the posterior Markov jump process. The time-independent rates of the prior process are in contrast trained akin to generative adversarial networks. We test our approach on synthetic data sampled from ground-truth Markov jump processes, experimental switching ion channel data and molecular dynamics simulations. Source code to reproduce our experiments is available online.
☆ Reliable Off-Policy Learning for Dosage Combinations
Decision-making in personalized medicine such as cancer therapy or critical care must often make choices for dosage combinations, i.e., multiple continuous treatments. Existing work for this task has modeled the effect of multiple treatments independently, while estimating the joint effect has received little attention but comes with non-trivial challenges. In this paper, we propose a novel method for reliable off-policy learning for dosage combinations. Our method proceeds along three steps: (1) We develop a tailored neural network that estimates the individualized dose-response function while accounting for the joint effect of multiple dependent dosages. (2) We estimate the generalized propensity score using conditional normalizing flows in order to detect regions with limited overlap in the shared covariate-treatment space. (3) We present a gradient-based learning algorithm to find the optimal, individualized dosage combinations. Here, we ensure reliable estimation of the policy value by avoiding regions with limited overlap. We finally perform an extensive evaluation of our method to show its effectiveness. To the best of our knowledge, ours is the first work to provide a method for reliable off-policy learning for optimal dosage combinations.
☆ Bures-Wasserstein Means of Graphs
Finding the mean of sampled data is a fundamental task in machine learning and statistics. However, in cases where the data samples are graph objects, defining a mean is an inherently difficult task. We propose a novel framework for defining a graph mean via embeddings in the space of smooth graph signal distributions, where graph similarity can be measured using the Wasserstein metric. By finding a mean in this embedding space, we can recover a mean graph that preserves structural information. We establish the existence and uniqueness of the novel graph mean, and provide an iterative algorithm for computing it. To highlight the potential of our framework as a valuable tool for practical applications in machine learning, it is evaluated on various tasks, including k-means clustering of structured graphs, classification of functional brain networks, and semi-supervised node classification in multi-layer graphs. Our experimental results demonstrate that our approach achieves consistent performance, outperforms existing baseline approaches, and improves state-of-the-art methods.
☆ APPRAISER: DNN Fault Resilience Analysis Employing Approximation Errors
Nowadays, the extensive exploitation of Deep Neural Networks (DNNs) in safety-critical applications raises new reliability concerns. In practice, methods for fault injection by emulation in hardware are efficient and widely used to study the resilience of DNN architectures for mitigating reliability issues already at the early design stages. However, the state-of-the-art methods for fault injection by emulation incur a spectrum of time-, design- and control-complexity problems. To overcome these issues, a novel resiliency assessment method called APPRAISER is proposed that applies functional approximation for a non-conventional purpose and employs approximate computing errors for its interest. By adopting this concept in the resiliency assessment domain, APPRAISER provides thousands of times speed-up in the assessment process, while keeping high accuracy of the analysis. In this paper, APPRAISER is validated by comparing it with state-of-the-art approaches for fault injection by emulation in FPGA. By this, the feasibility of the idea is demonstrated, and a new perspective in resiliency evaluation for DNNs is opened.
comment: 5 pages, 2 tables, 6 figures
☆ Data Representations' Study of Latent Image Manifolds ICML 2023
Deep neural networks have been demonstrated to achieve phenomenal success in many domains, and yet their inner mechanisms are not well understood. In this paper, we investigate the curvature of image manifolds, i.e., the manifold deviation from being flat in its principal directions. We find that state-of-the-art trained convolutional neural networks for image classification have a characteristic curvature profile along layers: an initial steep increase, followed by a long phase of a plateau, and followed by another increase. In contrast, this behavior does not appear in untrained networks in which the curvature flattens. We also show that the curvature gap between the last two layers has a strong correlation with the generalization capability of the network. Moreover, we find that the intrinsic dimension of latent codes is not necessarily indicative of curvature. Finally, we observe that common regularization methods such as mixup yield flatter representations when compared to other methods. Our experiments show consistent results over a variety of deep learning architectures and multiple data sets. Our code is publicly available at https://github.com/azencot-group/CRLM
comment: Accepted to ICML 2023
☆ Unbalanced Low-rank Optimal Transport Solvers
The relevance of optimal transport methods to machine learning has long been hindered by two salient limitations. First, the $O(n^3)$ computational cost of standard sample-based solvers (when used on batches of $n$ samples) is prohibitive. Second, the mass conservation constraint makes OT solvers too rigid in practice: because they must match \textit{all} points from both measures, their output can be heavily influenced by outliers. A flurry of recent works in OT has addressed these computational and modelling limitations, but has resulted in two separate strains of methods: While the computational outlook was much improved by entropic regularization, more recent $O(n)$ linear-time \textit{low-rank} solvers hold the promise to scale up OT further. On the other hand, modelling rigidities have been eased owing to unbalanced variants of OT, that rely on penalization terms to promote, rather than impose, mass conservation. The goal of this paper is to merge these two strains, to achieve the promise of \textit{both} versatile/scalable unbalanced/low-rank OT solvers. We propose custom algorithms to implement these extensions for the linear OT problem and its Fused-Gromov-Wasserstein generalization, and demonstrate their practical relevance to challenging spatial transcriptomics matching problems.
☆ Learning Representations without Compositional Assumptions
This paper addresses unsupervised representation learning on tabular data containing multiple views generated by distinct sources of measurement. Traditional methods, which tackle this problem using the multi-view framework, are constrained by predefined assumptions that assume feature sets share the same information and representations should learn globally shared factors. However, this assumption is not always valid for real-world tabular datasets with complex dependencies between feature sets, resulting in localized information that is harder to learn. To overcome this limitation, we propose a data-driven approach that learns feature set dependencies by representing feature sets as graph nodes and their relationships as learnable edges. Furthermore, we introduce LEGATO, a novel hierarchical graph autoencoder that learns a smaller, latent graph to aggregate information from multiple views dynamically. This approach results in latent graph components that specialize in capturing localized information from different regions of the input, leading to superior downstream performance.
☆ A rule-general abductive learning by rough sets
In real-world tasks, there is usually a large amount of unlabeled data and labeled data. The task of combining the two to learn is known as semi-supervised learning. Experts can use logical rules to label unlabeled data, but this operation is costly. The combination of perception and reasoning has a good effect in processing such semi-supervised tasks with domain knowledge. However, acquiring domain knowledge and the correction, reduction and generation of rules remain complex problems to be solved. Rough set theory is an important method for solving knowledge processing in information systems. In this paper, we propose a rule general abductive learning by rough set (RS-ABL). By transforming the target concept and sub-concepts of rules into information tables, rough set theory is used to solve the acquisition of domain knowledge and the correction, reduction and generation of rules at a lower cost. This framework can also generate more extensive negative rules to enhance the breadth of the knowledge base. Compared with the traditional semi-supervised learning method, RS-ABL has higher accuracy in dealing with semi-supervised tasks.
☆ Is Rewiring Actually Helpful in Graph Neural Networks?
Graph neural networks compute node representations by performing multiple message-passing steps that consist in local aggregations of node features. Having deep models that can leverage longer-range interactions between nodes is hindered by the issues of over-smoothing and over-squashing. In particular, the latter is attributed to the graph topology which guides the message-passing, causing a node representation to become insensitive to information contained at distant nodes. Many graph rewiring methods have been proposed to remedy or mitigate this problem. However, properly evaluating the benefits of these methods is made difficult by the coupling of over-squashing with other issues strictly related to model training, such as vanishing gradients. Therefore, we propose an evaluation setting based on message-passing models that do not require training to compute node and graph representations. We perform a systematic experimental comparison on real-world node and graph classification tasks, showing that rewiring the underlying graph rarely does confer a practical benefit for message-passing.
☆ Red Teaming Language Model Detectors with Language Models
The prevalence and high capacity of large language models (LLMs) present significant safety and ethical risks when malicious users exploit them for automated content generation. To prevent the potentially deceptive usage of LLMs, recent works have proposed several algorithms to detect machine-generated text. In this paper, we systematically test the reliability of the existing detectors, by designing two types of attack strategies to fool the detectors: 1) replacing words with their synonyms based on the context; 2) altering the writing style of generated text. These strategies are implemented by instructing LLMs to generate synonymous word substitutions or writing directives that modify the style without human involvement, and the LLMs leveraged in the attack can also be protected by detectors. Our research reveals that our attacks effectively compromise the performance of all tested detectors, thereby underscoring the urgent need for the development of more robust machine-generated text detection systems.
comment: Work in progress. Zhouxing Shi, Yihan Wang and Fan Yin are ordered alphabetically
☆ Optimal Decision Trees for Separable Objectives: Pushing the Limits of Dynamic Programming
Global optimization of decision trees has shown to be promising in terms of accuracy, size, and consequently human comprehensibility. However, many of the methods used rely on general-purpose solvers for which scalability remains an issue. Dynamic programming methods have been shown to scale much better because they exploit the tree structure by solving subtrees as independent subproblems. However, this only works when an objective can be optimized separately for subtrees. We explore this relationship in detail and show necessary and sufficient conditions for such separability and generalize previous dynamic programming approaches into a framework that can optimize any combination of separable objectives and constraints. Experiments on four application domains show the general applicability of this framework, while outperforming the scalability of general-purpose solvers by a large margin.
☆ Investigation of the Robustness of Neural Density Fields
Recent advances in modeling density distributions, so-called neural density fields, can accurately describe the density distribution of celestial bodies without, e.g., requiring a shape model - properties of great advantage when designing trajectories close to these bodies. Previous work introduced this approach, but several open questions remained. This work investigates neural density fields and their relative errors in the context of robustness to external factors like noise or constraints during training, like the maximal available gravity signal strength due to a certain distance exemplified for 433 Eros and 67P/Churyumov-Gerasimenko. It is found that both models trained on a polyhedral and mascon ground truth perform similarly, indicating that the ground truth is not the accuracy bottleneck. The impact of solar radiation pressure on a typical probe affects training neglectable, with the relative error being of the same magnitude as without noise. However, limiting the precision of measurement data by applying Gaussian noise hurts the obtainable precision. Further, pretraining is shown as practical in order to speed up network training. Hence, this work demonstrates that training neural networks for the gravity inversion problem is appropriate as long as the gravity signal is distinguishable from noise. Code and results are available at https://github.com/gomezzz/geodesyNets
comment: Will be presented as a poster at the GNC 2023 conference
☆ An Efficient Machine Learning-based Channel Prediction Technique for OFDM Sub-Bands
The acquisition of accurate channel state information (CSI) is of utmost importance since it provides performance improvement of wireless communication systems. However, acquiring accurate CSI, which can be done through channel estimation or channel prediction, is an intricate task due to the complexity of the time-varying and frequency selectivity of the wireless environment. To this end, we propose an efficient machine learning (ML)-based technique for channel prediction in orthogonal frequency-division multiplexing (OFDM) sub-bands. The novelty of the proposed approach lies in the training of channel fading samples used to estimate future channel behaviour in selective fading.
♻ ☆ How to Sift Out a Clean Data Subset in the Presence of Data Poisoning?
Given the volume of data needed to train modern machine learning models, external suppliers are increasingly used. However, incorporating external data poses data poisoning risks, wherein attackers manipulate their data to degrade model utility or integrity. Most poisoning defenses presume access to a set of clean data (or base set). While this assumption has been taken for granted, given the fast-growing research on stealthy poisoning attacks, a question arises: can defenders really identify a clean subset within a contaminated dataset to support defenses? This paper starts by examining the impact of poisoned samples on defenses when they are mistakenly mixed into the base set. We analyze five defenses and find that their performance deteriorates dramatically with less than 1% poisoned points in the base set. These findings suggest that sifting out a base set with high precision is key to these defenses' performance. Motivated by these observations, we study how precise existing automated tools and human inspection are at identifying clean data in the presence of data poisoning. Unfortunately, neither effort achieves the precision needed. Worse yet, many of the outcomes are worse than random selection. In addition to uncovering the challenge, we propose a practical countermeasure, Meta-Sift. Our method is based on the insight that existing attacks' poisoned samples shifts from clean data distributions. Hence, training on the clean portion of a dataset and testing on the corrupted portion will result in high prediction loss. Leveraging the insight, we formulate a bilevel optimization to identify clean data and further introduce a suite of techniques to improve efficiency and precision. Our evaluation shows that Meta-Sift can sift a clean base set with 100% precision under a wide range of poisoning attacks. The selected base set is large enough to give rise to successful defenses.
comment: 13 pages of the main text
♻ ☆ RARR: Researching and Revising What Language Models Say, Using Language Models ACL 2023
Language models (LMs) now excel at many tasks such as few-shot learning, question answering, reasoning, and dialog. However, they sometimes generate unsupported or misleading content. A user cannot easily determine whether their outputs are trustworthy or not, because most LMs do not have any built-in mechanism for attribution to external evidence. To enable attribution while still preserving all the powerful advantages of recent generation models, we propose RARR (Retrofit Attribution using Research and Revision), a system that 1) automatically finds attribution for the output of any text generation model and 2) post-edits the output to fix unsupported content while preserving the original output as much as possible. When applied to the output of several state-of-the-art LMs on a diverse set of generation tasks, we find that RARR significantly improves attribution while otherwise preserving the original input to a much greater degree than previously explored edit models. Furthermore, the implementation of RARR requires only a handful of training examples, a large language model, and standard web search.
comment: ACL 2023
♻ ☆ Efficient and Degree-Guided Graph Generation via Discrete Diffusion Modeling ICML 2023
Diffusion-based generative graph models have been proven effective in generating high-quality small graphs. However, they need to be more scalable for generating large graphs containing thousands of nodes desiring graph statistics. In this work, we propose EDGE, a new diffusion-based generative graph model that addresses generative tasks with large graphs. To improve computation efficiency, we encourage graph sparsity by using a discrete diffusion process that randomly removes edges at each time step and finally obtains an empty graph. EDGE only focuses on a portion of nodes in the graph at each denoising step. It makes much fewer edge predictions than previous diffusion-based models. Moreover, EDGE admits explicitly modeling the node degrees of the graphs, further improving the model performance. The empirical study shows that EDGE is much more efficient than competing methods and can generate large graphs with thousands of nodes. It also outperforms baseline models in generation quality: graphs generated by our approach have more similar graph statistics to those of the training graphs.
comment: ICML 2023, camera-ready revision
♻ ☆ Dropout Reduces Underfitting ICML 2023
Introduced by Hinton et al. in 2012, dropout has stood the test of time as a regularizer for preventing overfitting in neural networks. In this study, we demonstrate that dropout can also mitigate underfitting when used at the start of training. During the early phase, we find dropout reduces the directional variance of gradients across mini-batches and helps align the mini-batch gradients with the entire dataset's gradient. This helps counteract the stochasticity of SGD and limit the influence of individual batches on model training. Our findings lead us to a solution for improving performance in underfitting models - early dropout: dropout is applied only during the initial phases of training, and turned off afterwards. Models equipped with early dropout achieve lower final training loss compared to their counterparts without dropout. Additionally, we explore a symmetric technique for regularizing overfitting models - late dropout, where dropout is not used in the early iterations and is only activated later in training. Experiments on ImageNet and various vision tasks demonstrate that our methods consistently improve generalization accuracy. Our results encourage more research on understanding regularization in deep learning and our methods can be useful tools for future neural network training, especially in the era of large data. Code is available at https://github.com/facebookresearch/dropout.
comment: ICML 2023
♻ ☆ Fast Yet Effective Machine Unlearning
Unlearning the data observed during the training of a machine learning (ML) model is an important task that can play a pivotal role in fortifying the privacy and security of ML-based applications. This paper raises the following questions: (i) can we unlearn a single or multiple class(es) of data from a ML model without looking at the full training data even once? (ii) can we make the process of unlearning fast and scalable to large datasets, and generalize it to different deep networks? We introduce a novel machine unlearning framework with error-maximizing noise generation and impair-repair based weight manipulation that offers an efficient solution to the above questions. An error-maximizing noise matrix is learned for the class to be unlearned using the original model. The noise matrix is used to manipulate the model weights to unlearn the targeted class of data. We introduce impair and repair steps for a controlled manipulation of the network weights. In the impair step, the noise matrix along with a very high learning rate is used to induce sharp unlearning in the model. Thereafter, the repair step is used to regain the overall performance. With very few update steps, we show excellent unlearning while substantially retaining the overall model accuracy. Unlearning multiple classes requires a similar number of update steps as for a single class, making our approach scalable to large problems. Our method is quite efficient in comparison to the existing methods, works for multi-class unlearning, does not put any constraints on the original optimization mechanism or network design, and works well in both small and large-scale vision tasks. This work is an important step towards fast and easy implementation of unlearning in deep networks. Source code: https://github.com/vikram2000b/Fast-Machine-Unlearning
comment: Accepted in IEEE Transactions on Neural Networks and Learning Systems
♻ ☆ Accuracy on the Curve: On the Nonlinear Correlation of ML Performance Between Data Subpopulations ICML 2023
Understanding the performance of machine learning (ML) models across diverse data distributions is critically important for reliable applications. Despite recent empirical studies positing a near-perfect linear correlation between in-distribution (ID) and out-of-distribution (OOD) accuracies, we empirically demonstrate that this correlation is more nuanced under subpopulation shifts. Through rigorous experimentation and analysis across a variety of datasets, models, and training epochs, we demonstrate that OOD performance often has a nonlinear correlation with ID performance in subpopulation shifts. Our findings, which contrast previous studies that have posited a linear correlation in model performance during distribution shifts, reveal a "moon shape" correlation (parabolic uptrend curve) between the test performance on the majority subpopulation and the minority subpopulation. This non-trivial nonlinear correlation holds across model architectures, hyperparameters, training durations, and the imbalance between subpopulations. Furthermore, we found that the nonlinearity of this "moon shape" is causally influenced by the degree of spurious correlations in the training data. Our controlled experiments show that stronger spurious correlation in the training data creates more nonlinear performance correlation. We provide complementary experimental and theoretical analyses for this phenomenon, and discuss its implications for ML reliability and fairness. Our work highlights the importance of understanding the nonlinear effects of model improvement on performance in different subpopulations, and has the potential to inform the development of more equitable and responsible machine learning models.
comment: Accepted to the main conference of ICML 2023
♻ ☆ Rethinking Counterfactual Explanations as Local and Regional Counterfactual Policies
Counterfactual Explanations (CE) face several unresolved challenges, such as ensuring stability, synthesizing multiple CEs, and providing plausibility and sparsity guarantees. From a more practical point of view, recent studies [Pawelczyk et al., 2022] show that the prescribed counterfactual recourses are often not implemented exactly by individuals and demonstrate that most state-of-the-art CE algorithms are very likely to fail in this noisy environment. To address these issues, we propose a probabilistic framework that gives a sparse local counterfactual rule for each observation, providing rules that give a range of values capable of changing decisions with high probability. These rules serve as a summary of diverse counterfactual explanations and yield robust recourses. We further aggregate these local rules into a regional counterfactual rule, identifying shared recourses for subgroups of the data. Our local and regional rules are derived from the Random Forest algorithm, which offers statistical guarantees and fidelity to data distribution by selecting recourses in high-density regions. Moreover, our rules are sparse as we first select the smallest set of variables having a high probability of changing the decision. We have conducted experiments to validate the effectiveness of our counterfactual rules in comparison to standard CE and recent similar attempts. Our methods are available as a Python package.
♻ ☆ Simple Disentanglement of Style and Content in Visual Representations ICML
Learning visual representations with interpretable features, i.e., disentangled representations, remains a challenging problem. Existing methods demonstrate some success but are hard to apply to large-scale vision datasets like ImageNet. In this work, we propose a simple post-processing framework to disentangle content and style in learned representations from pre-trained vision models. We model the pre-trained features probabilistically as linearly entangled combinations of the latent content and style factors and develop a simple disentanglement algorithm based on the probabilistic model. We show that the method provably disentangles content and style features and verify its efficacy empirically. Our post-processed features yield significant domain generalization performance improvements when the distribution shift occurs due to style changes or style-related spurious correlations.
comment: International Conference on Machine Learning (ICML) 2023
♻ ☆ Accurate Shapley Values for explaining tree-based models AISTATS
Shapley Values (SV) are widely used in explainable AI, but their estimation and interpretation can be challenging, leading to inaccurate inferences and explanations. As a starting point, we remind an invariance principle for SV and derive the correct approach for computing the SV of categorical variables that are particularly sensitive to the encoding used. In the case of tree-based models, we introduce two estimators of Shapley Values that exploit the tree structure efficiently and are more accurate than state-of-the-art methods. Simulations and comparisons are performed with state-of-the-art algorithms and show the practical gain of our approach. Finally, we discuss the limitations of Shapley Values as a local explanation. These methods are available as a Python package.
comment: Accepted at the 25th International Conference on Artificial Intelligence and Statistics (AISTATS), 2022. V2: The section on Active Shapley Values has been removed in this updated version
♻ ☆ Adaptive Conformal Prediction by Reweighting Nonconformity Score
Despite attractive theoretical guarantees and practical successes, Predictive Interval (PI) given by Conformal Prediction (CP) may not reflect the uncertainty of a given model. This limitation arises from CP methods using a constant correction for all test points, disregarding their individual uncertainties, to ensure coverage properties. To address this issue, we propose using a Quantile Regression Forest (QRF) to learn the distribution of nonconformity scores and utilizing the QRF's weights to assign more importance to samples with residuals similar to the test point. This approach results in PI lengths that are more aligned with the model's uncertainty. In addition, the weights learnt by the QRF provide a partition of the features space, allowing for more efficient computations and improved adaptiveness of the PI through groupwise conformalization. Our approach enjoys an assumption-free finite sample marginal and training-conditional coverage, and under suitable assumptions, it also ensures conditional coverage. Our methods work for any nonconformity score and are available as a Python package. We conduct experiments on simulated and real-world data that demonstrate significant improvements compared to existing methods.
♻ ☆ Zero-Shot Machine Unlearning
Modern privacy regulations grant citizens the right to be forgotten by products, services and companies. In case of machine learning (ML) applications, this necessitates deletion of data not only from storage archives but also from ML models. Due to an increasing need for regulatory compliance required for ML applications, machine unlearning is becoming an emerging research problem. The right to be forgotten requests come in the form of removal of a certain set or class of data from the already trained ML model. Practical considerations preclude retraining of the model from scratch after discarding the deleted data. The few existing studies use either the whole training data, or a subset of training data, or some metadata stored during training to update the model weights for unlearning. However, in many cases, no data related to the training process or training samples may be accessible for the unlearning purpose. We therefore ask the question: is it possible to achieve unlearning with zero training samples? In this paper, we introduce the novel problem of zero-shot machine unlearning that caters for the extreme but practical scenario where zero original data samples are available for use. We then propose two novel solutions for zero-shot machine unlearning based on (a) error minimizing-maximizing noise and (b) gated knowledge transfer. These methods remove the information of the forget data from the model while maintaining the model efficacy on the retain data. The zero-shot approach offers good protection against the model inversion attacks and membership inference attacks. We introduce a new evaluation metric, Anamnesis Index (AIN) to effectively measure the quality of the unlearning method. The experiments show promising results for unlearning in deep learning models on benchmark vision data-sets. The source code is available here: https://github.com/ayu987/zero-shot-unlearning
comment: Accepted in IEEE Transactions on Information Forensics and Security (TIFS)
♻ ☆ Cooperative Open-ended Learning Framework for Zero-shot Coordination
Zero-shot coordination in cooperative artificial intelligence (AI) remains a significant challenge, which means effectively coordinating with a wide range of unseen partners. Previous algorithms have attempted to address this challenge by optimizing fixed objectives within a population to improve strategy or behaviour diversity. However, these approaches can result in a loss of learning and an inability to cooperate with certain strategies within the population, known as cooperative incompatibility. To address this issue, we propose the Cooperative Open-ended LEarning (COLE) framework, which constructs open-ended objectives in cooperative games with two players from the perspective of graph theory to assess and identify the cooperative ability of each strategy. We further specify the framework and propose a practical algorithm that leverages knowledge from game theory and graph theory. Furthermore, an analysis of the learning process of the algorithm shows that it can efficiently overcome cooperative incompatibility. The experimental results in the Overcooked game environment demonstrate that our method outperforms current state-of-the-art methods when coordinating with different-level partners. Our demo is available at https://sites.google.com/view/cole-2023.
comment: 15 pages with 9 pages main body
♻ ☆ HUB: Guiding Learned Optimizers with Continuous Prompt Tuning
Learned optimizers are a crucial component of meta-learning. Recent advancements in scalable learned optimizers have demonstrated their superior performance over hand-designed optimizers in various tasks. However, certain characteristics of these models, such as an unstable learning curve, limited ability to handle unseen tasks and network architectures, difficult-to-control behaviours, and poor performance in fine-tuning tasks impede their widespread adoption. To tackle the issue of generalization in scalable learned optimizers, we propose a hybrid-update-based (HUB) optimization strategy inspired by recent advancements in hard prompt tuning and result selection techniques used in large language and vision models. This approach can be easily applied to any task that involves hand-designed or learned optimizer. By incorporating hand-designed optimizers as the second component in our hybrid approach, we are able to retain the benefits of learned optimizers while stabilizing the training process and, more importantly, improving testing performance. We validate our design through a total of 17 tasks, consisting of thirteen training from scratch and four fine-tuning settings. These tasks vary in model sizes, architectures, or dataset sizes, and the competing optimizers are hyperparameter-tuned. We outperform all competitors in 94% of the tasks with better testing performance. Furthermore, we conduct a theoretical analysis to examine the potential impact of our hybrid strategy on the behaviours and inherited traits of learned optimizers.
comment: Some table information is not accurate, author information not correct inside the pdf
♻ ☆ SNeRL: Semantic-aware Neural Radiance Fields for Reinforcement Learning ICML 2023
As previous representations for reinforcement learning cannot effectively incorporate a human-intuitive understanding of the 3D environment, they usually suffer from sub-optimal performances. In this paper, we present Semantic-aware Neural Radiance Fields for Reinforcement Learning (SNeRL), which jointly optimizes semantic-aware neural radiance fields (NeRF) with a convolutional encoder to learn 3D-aware neural implicit representation from multi-view images. We introduce 3D semantic and distilled feature fields in parallel to the RGB radiance fields in NeRF to learn semantic and object-centric representation for reinforcement learning. SNeRL outperforms not only previous pixel-based representations but also recent 3D-aware representations both in model-free and model-based reinforcement learning.
comment: ICML 2023. First two authors contributed equally. Order was determined by coin flip
♻ ☆ ZeroFlow: Fast Zero Label Scene Flow via Distillation
Scene flow estimation is the task of describing the 3D motion field between temporally successive point clouds. State-of-the-art methods use strong priors and test-time optimization techniques, but require on the order of tens of seconds for large-scale point clouds, making them unusable as computer vision primitives for real-time applications such as open world object detection. Feed forward methods are considerably faster, running on the order of tens to hundreds of milliseconds for large-scale point clouds, but require expensive human supervision. To address both limitations, we propose Scene Flow via Distillation, a simple distillation framework that uses a label-free optimization method to produce pseudo-labels to supervise a feed forward model. Our instantiation of this framework, ZeroFlow, produces scene flow estimates in real-time on large-scale point clouds at quality competitive with state-of-the-art methods while using zero human labels. Notably, at test-time ZeroFlow is over 1000$\times$ faster than label-free state-of-the-art optimization-based methods on large-scale point clouds and over 1000$\times$ cheaper to train on unlabeled data compared to the cost of human annotation of that data. To facilitate research reuse, we release our code, trained model weights, and high quality pseudo-labels for the Argoverse 2 and Waymo Open datasets.
comment: 9 pages, 4 pages of Supplemental
♻ ☆ Unlocking Slot Attention by Changing Optimal Transport Costs ICML
Slot attention is a powerful method for object-centric modeling in images and videos. However, its set-equivariance limits its ability to handle videos with a dynamic number of objects because it cannot break ties. To overcome this limitation, we first establish a connection between slot attention and optimal transport. Based on this new perspective we propose MESH (Minimize Entropy of Sinkhorn): a cross-attention module that combines the tiebreaking properties of unregularized optimal transport with the speed of regularized optimal transport. We evaluate slot attention using MESH on multiple object-centric learning benchmarks and find significant improvements over slot attention in every setting.
comment: Published at International Conference on Machine Learning (ICML) 2023
♻ ☆ Decepticons: Corrupted Transformers Breach Privacy in Federated Learning for Language Models ICLR 2023
A central tenet of Federated learning (FL), which trains models without centralizing user data, is privacy. However, previous work has shown that the gradient updates used in FL can leak user information. While the most industrial uses of FL are for text applications (e.g. keystroke prediction), nearly all attacks on FL privacy have focused on simple image classifiers. We propose a novel attack that reveals private user text by deploying malicious parameter vectors, and which succeeds even with mini-batches, multiple users, and long sequences. Unlike previous attacks on FL, the attack exploits characteristics of both the Transformer architecture and the token embedding, separately extracting tokens and positional embeddings to retrieve high-fidelity text. This work suggests that FL on text, which has historically been resistant to privacy attacks, is far more vulnerable than previously thought.
comment: First two authors contributed equally. Order chosen by coin flip. Published at ICLR 2023. Implementation available at github.com/JonasGeiping/breaching
♻ ☆ On the Forward Invariance of Neural ODEs ICML2023
We propose a new method to ensure neural ordinary differential equations (ODEs) satisfy output specifications by using invariance set propagation. Our approach uses a class of control barrier functions to transform output specifications into constraints on the parameters and inputs of the learning system. This setup allows us to achieve output specification guarantees simply by changing the constrained parameters/inputs both during training and inference. Moreover, we demonstrate that our invariance set propagation through data-controlled neural ODEs not only maintains generalization performance but also creates an additional degree of robustness by enabling causal manipulation of the system's parameters/inputs. We test our method on a series of representation learning tasks, including modeling physical dynamics and convexity portraits, as well as safe collision avoidance for autonomous vehicles.
comment: 25 pages, accepted in ICML2023, website: https://weixy21.github.io/invariance/
♻ ☆ Why Random Pruning Is All We Need to Start Sparse ICML
Random masks define surprisingly effective sparse neural network models, as has been shown empirically. The resulting sparse networks can often compete with dense architectures and state-of-the-art lottery ticket pruning algorithms, even though they do not rely on computationally expensive prune-train iterations and can be drawn initially without significant computational overhead. We offer a theoretical explanation of how random masks can approximate arbitrary target networks if they are wider by a logarithmic factor in the inverse sparsity $1 / \log(1/\text{sparsity})$. This overparameterization factor is necessary at least for 3-layer random networks, which elucidates the observed degrading performance of random networks at higher sparsity. At moderate to high sparsity levels, however, our results imply that sparser networks are contained within random source networks so that any dense-to-sparse training scheme can be turned into a computationally more efficient sparse-to-sparse one by constraining the search to a fixed random mask. We demonstrate the feasibility of this approach in experiments for different pruning methods and propose particularly effective choices of initial layer-wise sparsity ratios of the random source network. As a special case, we show theoretically and experimentally that random source networks also contain strong lottery tickets.
comment: Accepted for publication at ICML, 2023
♻ ☆ What can online reinforcement learning with function approximation benefit from general coverage conditions? ICML 2023
In online reinforcement learning (RL), instead of employing standard structural assumptions on Markov decision processes (MDPs), using a certain coverage condition (original from offline RL) is enough to ensure sample-efficient guarantees (Xie et al. 2023). In this work, we focus on this new direction by digging more possible and general coverage conditions, and study the potential and the utility of them in efficient online RL. We identify more concepts, including the $L^p$ variant of concentrability, the density ratio realizability, and trade-off on the partial/rest coverage condition, that can be also beneficial to sample-efficient online RL, achieving improved regret bound. Furthermore, if exploratory offline data are used, under our coverage conditions, both statistically and computationally efficient guarantees can be achieved for online RL. Besides, even though the MDP structure is given, e.g., linear MDP, we elucidate that, good coverage conditions are still beneficial to obtain faster regret bound beyond $\widetilde{O}(\sqrt{T})$ and even a logarithmic order regret. These results provide a good justification for the usage of general coverage conditions in efficient online RL.
comment: Accepted by ICML 2023
♻ ☆ Reinforcement Learning with Human Feedback: Learning Dynamic Choices via Pessimism
In this paper, we study offline Reinforcement Learning with Human Feedback (RLHF) where we aim to learn the human's underlying reward and the MDP's optimal policy from a set of trajectories induced by human choices. RLHF is challenging for multiple reasons: large state space but limited human feedback, the bounded rationality of human decisions, and the off-policy distribution shift. In this paper, we focus on the Dynamic Discrete Choice (DDC) model for modeling and understanding human choices. DCC, rooted in econometrics and decision theory, is widely used to model a human decision-making process with forward-looking and bounded rationality. We propose a \underline{D}ynamic-\underline{C}hoice-\underline{P}essimistic-\underline{P}olicy-\underline{O}ptimization (DCPPO) method. \ The method involves a three-stage process: The first step is to estimate the human behavior policy and the state-action value function via maximum likelihood estimation (MLE); the second step recovers the human reward function via minimizing Bellman mean squared error using the learned value functions; the third step is to plug in the learned reward and invoke pessimistic value iteration for finding a near-optimal policy. With only single-policy coverage (i.e., optimal policy) of the dataset, we prove that the suboptimality of DCPPO almost matches the classical pessimistic offline RL algorithm in terms of suboptimality's dependency on distribution shift and dimension. To the best of our knowledge, this paper presents the first theoretical guarantees for off-policy offline RLHF with dynamic discrete choice model.
♻ ☆ On Balancing Bias and Variance in Unsupervised Multi-Source-Free Domain Adaptation ICML 2023
Due to privacy, storage, and other constraints, there is a growing need for unsupervised domain adaptation techniques in machine learning that do not require access to the data used to train a collection of source models. Existing methods for multi-source-free domain adaptation (MSFDA) typically train a target model using pseudo-labeled data produced by the source models, which focus on improving the pseudo-labeling techniques or proposing new training objectives. Instead, we aim to analyze the fundamental limits of MSFDA. In particular, we develop an information-theoretic bound on the generalization error of the resulting target model, which illustrates an inherent bias-variance trade-off. We then provide insights on how to balance this trade-off from three perspectives, including domain aggregation, selective pseudo-labeling, and joint feature alignment, which leads to the design of novel algorithms. Experiments on multiple datasets validate our theoretical analysis and demonstrate the state-of-art performance of the proposed algorithm, especially on some of the most challenging datasets, including Office-Home and DomainNet.
comment: ICML 2023
♻ ☆ Attend-and-Excite: Attention-Based Semantic Guidance for Text-to-Image Diffusion Models SIGGRAPH 2023
Recent text-to-image generative models have demonstrated an unparalleled ability to generate diverse and creative imagery guided by a target text prompt. While revolutionary, current state-of-the-art diffusion models may still fail in generating images that fully convey the semantics in the given text prompt. We analyze the publicly available Stable Diffusion model and assess the existence of catastrophic neglect, where the model fails to generate one or more of the subjects from the input prompt. Moreover, we find that in some cases the model also fails to correctly bind attributes (e.g., colors) to their corresponding subjects. To help mitigate these failure cases, we introduce the concept of Generative Semantic Nursing (GSN), where we seek to intervene in the generative process on the fly during inference time to improve the faithfulness of the generated images. Using an attention-based formulation of GSN, dubbed Attend-and-Excite, we guide the model to refine the cross-attention units to attend to all subject tokens in the text prompt and strengthen - or excite - their activations, encouraging the model to generate all subjects described in the text prompt. We compare our approach to alternative approaches and demonstrate that it conveys the desired concepts more faithfully across a range of text prompts.
comment: Accepted to SIGGRAPH 2023; Project page available at https://yuval-alaluf.github.io/Attend-and-Excite/
♻ ☆ What Can Be Learnt With Wide Convolutional Neural Networks?
Understanding how convolutional neural networks (CNNs) can efficiently learn high-dimensional functions remains a fundamental challenge. A popular belief is that these models harness the local and hierarchical structure of natural data such as images. Yet, we lack a quantitative understanding of how such structure affects performance, e.g., the rate of decay of the generalisation error with the number of training samples. In this paper, we study infinitely-wide deep CNNs in the kernel regime. First, we show that the spectrum of the corresponding kernel inherits the hierarchical structure of the network, and we characterise its asymptotics. Then, we use this result together with generalisation bounds to prove that deep CNNs adapt to the spatial scale of the target function. In particular, we find that if the target function depends on low-dimensional subsets of adjacent input variables, then the decay of the error is controlled by the effective dimensionality of these subsets. Conversely, if the target function depends on the full set of input variables, then the error decay is controlled by the input dimension. We conclude by computing the generalisation error of a deep CNN trained on the output of another deep CNN with randomly-initialised parameters. Interestingly, we find that, despite their hierarchical structure, the functions generated by infinitely-wide deep CNNs are too rich to be efficiently learnable in high dimension.
♻ ☆ Explaining the effects of non-convergent sampling in the training of Energy-Based Models ICML 2023
In this paper, we quantify the impact of using non-convergent Markov chains to train Energy-Based models (EBMs). In particular, we show analytically that EBMs trained with non-persistent short runs to estimate the gradient can perfectly reproduce a set of empirical statistics of the data, not at the level of the equilibrium measure, but through a precise dynamical process. Our results provide a first-principles explanation for the observations of recent works proposing the strategy of using short runs starting from random initial conditions as an efficient way to generate high-quality samples in EBMs, and lay the groundwork for using EBMs as diffusion models. After explaining this effect in generic EBMs, we analyze two solvable models in which the effect of the non-convergent sampling in the trained parameters can be described in detail. Finally, we test these predictions numerically on a ConvNet EBM and a Boltzmann machine.
comment: Accepted at ICML 2023
♻ ☆ Generalizable Memory-driven Transformer for Multivariate Long Sequence Time-series Forecasting
Multivariate long sequence time-series forecasting (M-LSTF) is a practical but challenging problem. Unlike traditional timer-series forecasting tasks, M-LSTF tasks are more challenging from two aspects: 1) M-LSTF models need to learn time-series patterns both within and between multiple time features; 2) Under the rolling forecasting setting, the similarity between two consecutive training samples increases with the increasing prediction length, which makes models more prone to overfitting. In this paper, we propose a generalizable memory-driven Transformer to target M-LSTF problems. Specifically, we first propose a global-level memory component to drive the forecasting procedure by integrating multiple time-series features. In addition, we adopt a progressive fashion to train our model to increase its generalizability, in which we gradually introduce Bernoulli noises to training samples. Extensive experiments have been performed on five different datasets across multiple fields. Experimental results demonstrate that our approach can be seamlessly plugged into varying Transformer-based models to improve their performances up to roughly 30%. Particularly, this is the first work to specifically focus on the M-LSTF tasks to the best of our knowledge.
comment: Tech report
♻ ☆ A neural network-supported two-stage algorithm for lightweight dereverberation on hearing devices
A two-stage lightweight online dereverberation algorithm for hearing devices is presented in this paper. The approach combines a multi-channel multi-frame linear filter with a single-channel single-frame post-filter. Both components rely on power spectral density (PSD) estimates provided by deep neural networks (DNNs). By deriving new metrics analyzing the dereverberation performance in various time ranges, we confirm that directly optimizing for a criterion at the output of the multi-channel linear filtering stage results in a more efficient dereverberation as compared to placing the criterion at the output of the DNN to optimize the PSD estimation. More concretely, we show that training this stage end-to-end helps further remove the reverberation in the range accessible to the filter, thus increasing the \textit{early-to-moderate} reverberation ratio. We argue and demonstrate that it can then be well combined with a post-filtering stage to efficiently suppress the residual late reverberation, thereby increasing the \textit{early-to-final} reverberation ratio. This proposed two stage procedure is shown to be both very effective in terms of dereverberation performance and computational demands, as compared to e.g. recent state-of-the-art DNN approaches. Furthermore, the proposed two-stage system can be adapted to the needs of different types of hearing-device users by controlling the amount of reduction of early reflections.
comment: Accepted for publication in EURASIP Journal on Audio, Speech and Music Processing
♻ ☆ Computationally Efficient 3D MRI Reconstruction with Adaptive MLP MICCAI 2023
Compared with 2D MRI, 3D MRI provides superior volumetric spatial resolution and signal-to-noise ratio. However, it is more challenging to reconstruct 3D MRI images. Current methods are mainly based on convolutional neural networks (CNN) with small kernels, which are difficult to scale up to have sufficient fitting power for 3D MRI reconstruction due to the large image size and GPU memory constraint. Furthermore, MRI reconstruction is a deconvolution problem, which demands long-distance information that is difficult to capture by CNNs with small convolution kernels. The multi-layer perceptron (MLP) can model such long-distance information, but it requires a fixed input size. In this paper, we proposed Recon3DMLP, a hybrid of CNN modules with small kernels for low-frequency reconstruction and adaptive MLP (dMLP) modules with large kernels to boost the high-frequency reconstruction, for 3D MRI reconstruction. We further utilized the circular shift operation based on MRI physics such that dMLP accepts arbitrary image size and can extract global information from the entire FOV. We also propose a GPU memory efficient data fidelity module that can reduce $>$50$\%$ memory. We compared Recon3DMLP with other CNN-based models on a high-resolution (HR) 3D MRI dataset. Recon3DMLP improves HR 3D reconstruction and outperforms several existing CNN-based models under similar GPU memory consumption, which demonstrates that Recon3DMLP is a practical solution for HR 3D MRI reconstruction.
comment: MICCAI 2023 early accept
♻ ☆ On Differentially Private Federated Linear Contextual Bandits
We consider cross-silo federated linear contextual bandit (LCB) problem under differential privacy, where multiple silos (agents) interact with the local users and communicate via a central server to realize collaboration while without sacrificing each user's privacy. We identify three issues in the state-of-the-art: (i) failure of claimed privacy protection and (ii) incorrect regret bound due to noise miscalculation and (iii) ungrounded communication cost. To resolve these issues, we take a two-step principled approach. First, we design an algorithmic framework consisting of a generic federated LCB algorithm and flexible privacy protocols. Then, leveraging the proposed framework, we study federated LCBs under two different privacy constraints. We first establish privacy and regret guarantees under silo-level local differential privacy, which fix the issues present in state-of-the-art algorithm. To further improve the regret performance, we next consider shuffle model of differential privacy, under which we show that our algorithm can achieve nearly ``optimal'' regret without a trusted server. We accomplish this via two different schemes -- one relies on a new result on privacy amplification via shuffling for DP mechanisms and another one leverages the integration of a shuffle protocol for vector sum into the tree-based mechanism, both of which might be of independent interest. Finally, we support our theoretical results with numerical evaluations over contextual bandit instances generated from both synthetic and real-life data.
♻ ☆ Can Self-Supervised Neural Representations Pre-Trained on Human Speech distinguish Animal Callers?
Self-supervised learning (SSL) models use only the intrinsic structure of a given signal, independent of its acoustic domain, to extract essential information from the input to an embedding space. This implies that the utility of such representations is not limited to modeling human speech alone. Building on this understanding, this paper explores the cross-transferability of SSL neural representations learned from human speech to analyze bio-acoustic signals. We conduct a caller discrimination analysis and a caller detection study on Marmoset vocalizations using eleven SSL models pre-trained with various pretext tasks. The results show that the embedding spaces carry meaningful caller information and can successfully distinguish the individual identities of Marmoset callers without fine-tuning. This demonstrates that representations pre-trained on human speech can be effectively applied to the bio-acoustics domain, providing valuable insights for future investigations in this field.
comment: Accepted at Interspeech 2023
♻ ☆ The Stable Artist: Steering Semantics in Diffusion Latent Space
Large, text-conditioned generative diffusion models have recently gained a lot of attention for their impressive performance in generating high-fidelity images from text alone. However, achieving high-quality results is almost unfeasible in a one-shot fashion. On the contrary, text-guided image generation involves the user making many slight changes to inputs in order to iteratively carve out the envisioned image. However, slight changes to the input prompt often lead to entirely different images being generated, and thus the control of the artist is limited in its granularity. To provide flexibility, we present the Stable Artist, an image editing approach enabling fine-grained control of the image generation process. The main component is semantic guidance (SEGA) which steers the diffusion process along variable numbers of semantic directions. This allows for subtle edits to images, changes in composition and style, as well as optimization of the overall artistic conception. Furthermore, SEGA enables probing of latent spaces to gain insights into the representation of concepts learned by the model, even complex ones such as 'carbon emission'. We demonstrate the Stable Artist on several tasks, showcasing high-quality image editing and composition.
comment: This is a report of preliminary results. A full version of the paper is available at: arXiv:2301.12247
♻ ☆ ILLUME: Rationalizing Vision-Language Models through Human Interactions ICML
Bootstrapping from pre-trained language models has been proven to be an efficient approach for building vision-language models (VLM) for tasks such as image captioning or visual question answering. However, outputs of these models rarely align with user's rationales for specific answers. In order to improve this alignment and reinforce commonsense reasons, we propose a tuning paradigm based on human interactions with machine-generated data. Our ILLUME executes the following loop: Given an image-question-answer prompt, the VLM samples multiple candidate rationales, and a human critic provides feedback via preference selection, used for fine-tuning. This loop increases the training data and gradually carves out the VLM's rationalization capabilities that are aligned with human intent. Our exhaustive experiments demonstrate that ILLUME is competitive with standard supervised finetuning while using significantly fewer training data and only requiring minimal feedback.
comment: Proceedings of the 40th International Conference on Machine Learning (ICML), 2023
♻ ☆ Can We Scale Transformers to Predict Parameters of Diverse ImageNet Models? ICML 2023
Pretraining a neural network on a large dataset is becoming a cornerstone in machine learning that is within the reach of only a few communities with large-resources. We aim at an ambitious goal of democratizing pretraining. Towards that goal, we train and release a single neural network that can predict high quality ImageNet parameters of other neural networks. By using predicted parameters for initialization we are able to boost training of diverse ImageNet models available in PyTorch. When transferred to other datasets, models initialized with predicted parameters also converge faster and reach competitive final performance.
comment: ICML 2023, camera ready (7 tables with extra results added), code and models are at https://github.com/SamsungSAILMontreal/ghn3
♻ ☆ X-Paste: Revisiting Scalable Copy-Paste for Instance Segmentation using CLIP and StableDiffusion ICML 2023
Copy-Paste is a simple and effective data augmentation strategy for instance segmentation. By randomly pasting object instances onto new background images, it creates new training data for free and significantly boosts the segmentation performance, especially for rare object categories. Although diverse, high-quality object instances used in Copy-Paste result in more performance gain, previous works utilize object instances either from human-annotated instance segmentation datasets or rendered from 3D object models, and both approaches are too expensive to scale up to obtain good diversity. In this paper, we revisit Copy-Paste at scale with the power of newly emerged zero-shot recognition models (e.g., CLIP) and text2image models (e.g., StableDiffusion). We demonstrate for the first time that using a text2image model to generate images or zero-shot recognition model to filter noisily crawled images for different object categories is a feasible way to make Copy-Paste truly scalable. To make such success happen, we design a data acquisition and processing framework, dubbed ``X-Paste", upon which a systematic study is conducted. On the LVIS dataset, X-Paste provides impressive improvements over the strong baseline CenterNet2 with Swin-L as the backbone. Specifically, it archives +2.6 box AP and +2.1 mask AP gains on all classes and even more significant gains with +6.8 box AP, +6.5 mask AP on long-tail classes. Our code and models are available at https://github.com/yoctta/XPaste.
comment: ICML 2023, code is available at https://github.com/yoctta/XPaste
♻ ☆ Personalized Algorithmic Recourse with Preference Elicitation
Algorithmic Recourse (AR) is the problem of computing a sequence of actions that -- once performed by a user -- overturns an undesirable machine decision. It is paramount that the sequence of actions does not require too much effort for users to implement. Yet, most approaches to AR assume that actions cost the same for all users, and thus may recommend unfairly expensive recourse plans to certain users. Prompted by this observation, we introduce PEAR, the first human-in-the-loop approach capable of providing personalized algorithmic recourse tailored to the needs of any end-user. PEAR builds on insights from Bayesian Preference Elicitation to iteratively refine an estimate of the costs of actions by asking choice set queries to the target user. The queries themselves are computed by maximizing the Expected Utility of Selection, a principled measure of information gain accounting for uncertainty on both the cost estimate and the user's responses. PEAR integrates elicitation into a Reinforcement Learning agent coupled with Monte Carlo Tree Search to quickly identify promising recourse plans. Our empirical evaluation on real-world datasets highlights how PEAR produces high-quality personalized recourse in only a handful of iterations.
♻ ☆ DoG is SGD's Best Friend: A Parameter-Free Dynamic Step Size Schedule ICML 2023
We propose a tuning-free dynamic SGD step size formula, which we call Distance over Gradients (DoG). The DoG step sizes depend on simple empirical quantities (distance from the initial point and norms of gradients) and have no ``learning rate'' parameter. Theoretically, we show that a slight variation of the DoG formula enjoys strong parameter-free convergence guarantees for stochastic convex optimization assuming only \emph{locally bounded} stochastic gradients. Empirically, we consider a broad range of vision and language transfer learning tasks, and show that DoG's performance is close to that of SGD with tuned learning rate. We also propose a per-layer variant of DoG that generally outperforms tuned SGD, approaching the performance of tuned Adam. A PyTorch implementation is available at https://github.com/formll/dog
comment: To appear at ICML 2023
♻ ☆ Shedding a PAC-Bayesian Light on Adaptive Sliced-Wasserstein Distances
The Sliced-Wasserstein distance (SW) is a computationally efficient and theoretically grounded alternative to the Wasserstein distance. Yet, the literature on its statistical properties -- or, more accurately, its generalization properties -- with respect to the distribution of slices, beyond the uniform measure, is scarce. To bring new contributions to this line of research, we leverage the PAC-Bayesian theory and a central observation that SW may be interpreted as an average risk, the quantity PAC-Bayesian bounds have been designed to characterize. We provide three types of results: i) PAC-Bayesian generalization bounds that hold on what we refer as adaptive Sliced-Wasserstein distances, i.e. SW defined with respect to arbitrary distributions of slices (among which data-dependent distributions), ii) a principled procedure to learn the distribution of slices that yields maximally discriminative SW, by optimizing our theoretical bounds, and iii) empirical illustrations of our theoretical findings.
Reverse Engineering Self-Supervised Learning
Self-supervised learning (SSL) is a powerful tool in machine learning, but understanding the learned representations and their underlying mechanisms remains a challenge. This paper presents an in-depth empirical analysis of SSL-trained representations, encompassing diverse models, architectures, and hyperparameters. Our study reveals an intriguing aspect of the SSL training process: it inherently facilitates the clustering of samples with respect to semantic labels, which is surprisingly driven by the SSL objective's regularization term. This clustering process not only enhances downstream classification but also compresses the data information. Furthermore, we establish that SSL-trained representations align more closely with semantic classes rather than random classes. Remarkably, we show that learned representations align with semantic classes across various hierarchical levels, and this alignment increases during training and when moving deeper into the network. Our findings provide valuable insights into SSL's representation learning mechanisms and their impact on performance across different sets of classes.
♻ ☆ Learning the Dynamics of Sparsely Observed Interacting Systems ICML 2023
We address the problem of learning the dynamics of an unknown non-parametric system linking a target and a feature time series. The feature time series is measured on a sparse and irregular grid, while we have access to only a few points of the target time series. Once learned, we can use these dynamics to predict values of the target from the previous values of the feature time series. We frame this task as learning the solution map of a controlled differential equation (CDE). By leveraging the rich theory of signatures, we are able to cast this non-linear problem as a high-dimensional linear regression. We provide an oracle bound on the prediction error which exhibits explicit dependencies on the individual-specific sampling schemes. Our theoretical results are illustrated by simulations which show that our method outperforms existing algorithms for recovering the full time series while being computationally cheap. We conclude by demonstrating its potential on real-world epidemiological data.
comment: ICML 2023
♻ ☆ Elixir: Train a Large Language Model on a Small GPU Cluster
In recent years, large language models have achieved great success due to their unprecedented size. However, training these models poses a challenge for most researchers as it requires a substantial number of GPUs. To reduce GPU memory usage, memory partitioning, and memory offloading have been proposed. These approaches eliminate memory redundancies and offload memory usage to the CPU and NVMe memory, respectively, enabling training on small GPU clusters. However, directly deploying these solutions often leads to suboptimal efficiency. Only experienced experts can unleash the full potential of hardware by carefully tuning the distributed configuration. Thus, we present a novel solution, Elixir, which automates efficient large-model training based on pre-runtime model profiling. Elixir aims to identify the optimal combination of partitioning and offloading techniques to maximize training throughput. In our experiments, Elixir significantly outperforms the current state-of-the-art baseline. Our optimal configuration achieves up to a 3.4$\times$ speedup on GPT-2 models compared with SOTA solutions. We hope that our work will benefit individuals who lack computing resources and expertise, granting them access to large models. The beta version of Elixir is now available at https://github.com/hpcaitech/ColossalAI/tree/feature/elixir.
♻ ☆ Static Scheduling with Predictions Learned through Efficient Exploration ICML 2023
We study single-machine scheduling of jobs, each belonging to a job type that determines its duration distribution. We start by analyzing the scenario where the type characteristics are known and then move to two learning scenarios where the types are unknown: non-preemptive problems, where each started job must be completed before moving to another job; and preemptive problems, where job execution can be paused in the favor of moving to a different job. In both cases, we design algorithms that achieve sublinear excess cost, compared to the performance with known types, and prove lower bounds for the non-preemptive case. Notably, we demonstrate, both theoretically and through simulations, how preemptive algorithms can greatly outperform non-preemptive ones when the durations of different job types are far from one another, a phenomenon that does not occur when the type durations are known.
comment: Accepted to ICML 2023
♻ ☆ FusionRetro: Molecule Representation Fusion via In-Context Learning for Retrosynthetic Planning ICML 2023
Retrosynthetic planning aims to devise a complete multi-step synthetic route from starting materials to a target molecule. Current strategies use a decoupled approach of single-step retrosynthesis models and search algorithms, taking only the product as the input to predict the reactants for each planning step and ignoring valuable context information along the synthetic route. In this work, we propose a novel framework that utilizes context information for improved retrosynthetic planning. We view synthetic routes as reaction graphs and propose to incorporate context through three principled steps: encode molecules into embeddings, aggregate information over routes, and readout to predict reactants. Our approach is the first attempt to utilize in-context learning for retrosynthesis prediction in retrosynthetic planning. The entire framework can be efficiently optimized in an end-to-end fashion and produce more practical and accurate predictions. Comprehensive experiments demonstrate that by fusing in the context information over routes, our model significantly improves the performance of retrosynthetic planning over baselines that are not context-aware, especially for long synthetic routes. Code is available at https://github.com/SongtaoLiu0823/FusionRetro.
comment: Accepted by ICML 2023
♻ ☆ FedST: Secure Federated Shapelet Transformation for Time Series Classification
This paper explores how to customize time series classification (TSC) methods with the help of external data in a privacy-preserving federated learning (FL) scenario. To the best of our knowledge, we are the first to study on this essential topic. Achieving this goal requires us to seamlessly integrate the techniques from multiple fields including Data Mining, Machine Learning, and Security. In this paper, we systematically investigate existing TSC solutions for the centralized scenario and propose FedST, a novel FL-enabled TSC framework based on a shapelet transformation method. We recognize the federated shapelet search step as the kernel of FedST. Thus, we design a basic protocol for the FedST kernel that we prove to be secure and accurate. However, we identify that the basic protocol suffers from efficiency bottlenecks and the centralized acceleration techniques lose their efficacy due to the security issues. To speed up the federated protocol with security guarantee, we propose several optimizations tailored for the FL setting. Our theoretical analysis shows that the proposed methods are secure and more efficient. We conduct extensive experiments using both synthetic and real-world datasets. Empirical results show that our FedST solution is effective in terms of TSC accuracy, and the proposed optimizations can achieve three orders of magnitude of speedup.
♻ ☆ IB-RAR: Information Bottleneck as Regularizer for Adversarial Robustness
In this paper, we propose a novel method, IB-RAR, which uses Information Bottleneck (IB) to strengthen adversarial robustness for both adversarial training and non-adversarial-trained methods. We first use the IB theory to build regularizers as learning objectives in the loss function. Then, we filter out unnecessary features of intermediate representation according to their mutual information (MI) with labels, as the network trained with IB provides easily distinguishable MI for its features. Experimental results show that our method can be naturally combined with adversarial training and provides consistently better accuracy on new adversarial examples. Our method improves the accuracy by an average of 3.07% against five adversarial attacks for the VGG16 network, trained with three adversarial training benchmarks and the CIFAR-10 dataset. In addition, our method also provides good robustness for undefended methods, such as training with cross-entropy loss only. Finally, in the absence of adversarial training, the VGG16 network trained using our method and the CIFAR-10 dataset reaches an accuracy of 35.86% against PGD examples, while using all layers reaches 25.61% accuracy.
♻ ☆ Provably Convergent Schrödinger Bridge with Applications to Probabilistic Time Series Imputation ICML 2023
The Schr\"odinger bridge problem (SBP) is gaining increasing attention in generative modeling and showing promising potential even in comparison with the score-based generative models (SGMs). SBP can be interpreted as an entropy-regularized optimal transport problem, which conducts projections onto every other marginal alternatingly. However, in practice, only approximated projections are accessible and their convergence is not well understood. To fill this gap, we present a first convergence analysis of the Schr\"odinger bridge algorithm based on approximated projections. As for its practical applications, we apply SBP to probabilistic time series imputation by generating missing values conditioned on observed data. We show that optimizing the transport cost improves the performance and the proposed algorithm achieves the state-of-the-art result in healthcare and environmental data while exhibiting the advantage of exploring both temporal and feature patterns in probabilistic time series imputation.
comment: Accepted by ICML 2023
♻ ☆ Image Restoration with Mean-Reverting Stochastic Differential Equations ICML 2023
This paper presents a stochastic differential equation (SDE) approach for general-purpose image restoration. The key construction consists in a mean-reverting SDE that transforms a high-quality image into a degraded counterpart as a mean state with fixed Gaussian noise. Then, by simulating the corresponding reverse-time SDE, we are able to restore the origin of the low-quality image without relying on any task-specific prior knowledge. Crucially, the proposed mean-reverting SDE has a closed-form solution, allowing us to compute the ground truth time-dependent score and learn it with a neural network. Moreover, we propose a maximum likelihood objective to learn an optimal reverse trajectory that stabilizes the training and improves the restoration results. The experiments show that our proposed method achieves highly competitive performance in quantitative comparisons on image deraining, deblurring, and denoising, setting a new state-of-the-art on two deraining datasets. Finally, the general applicability of our approach is further demonstrated via qualitative results on image super-resolution, inpainting, and dehazing. Code is available at https://github.com/Algolzw/image-restoration-sde.
comment: Accepted by ICML 2023; Project page: https://algolzw.github.io/ir-sde/index.html
♻ ☆ Hierarchical Policy Blending as Inference for Reactive Robot Control ICRA 2023
Motion generation in cluttered, dense, and dynamic environments is a central topic in robotics, rendered as a multi-objective decision-making problem. Current approaches trade-off between safety and performance. On the one hand, reactive policies guarantee fast response to environmental changes at the risk of suboptimal behavior. On the other hand, planning-based motion generation provides feasible trajectories, but the high computational cost may limit the control frequency and thus safety. To combine the benefits of reactive policies and planning, we propose a hierarchical motion generation method. Moreover, we adopt probabilistic inference methods to formalize the hierarchical model and stochastic optimization. We realize this approach as a weighted product of stochastic, reactive expert policies, where planning is used to adaptively compute the optimal weights over the task horizon. This stochastic optimization avoids local optima and proposes feasible reactive plans that find paths in cluttered and dense environments. Our extensive experimental study in planar navigation and 6DoF manipulation shows that our proposed hierarchical motion generation method outperforms both myopic reactive controllers and online re-planning methods.
comment: 8 pages, 5 figures, 1 table, accepted at ICRA 2023
♻ ☆ Conformal Regression in Calorie Prediction for Team Jumbo-Visma
UCI WorldTour races, the premier men's elite road cycling tour, are grueling events that put physical fitness and endurance of riders to the test. The coaches of Team Jumbo-Visma have long been responsible for predicting the energy needs of each rider of the Dutch team for every race on the calendar. Those must be estimated to ensure riders have the energy and resources necessary to maintain a high level of performance throughout a race. This task, however, is both time-consuming and challenging, as it requires precise estimates of race speed and power output. Traditionally, the approach to predicting energy needs has relied on judgement and experience of coaches, but this method has its limitations and often leads to inaccurate predictions. In this paper, we propose a new, more effective approach to predicting energy needs for cycling races. By predicting the speed and power with regression models, we provide the coaches with calorie needs estimates for each individual rider per stage instantly. In addition, we compare methods to quantify uncertainty using conformal prediction. The empirical analysis of the jackknife+, jackknife-minmax, jackknife-minmax-after-bootstrap, CV+, CV-minmax, conformalized quantile regression, and inductive conformal prediction methods in conformal prediction reveals that all methods achieve valid prediction intervals. All but minmax-based methods also produce produce sufficiently narrow prediction intervals for decision-making. Furthermore, methods computing prediction intervals of fixed size produce tighter intervals for low significance values. Among the methods computing intervals of varying length across the input space, inductive conformal prediction computes narrower prediction intervals at larger significance level.
comment: 11 pages, 5 figures
♻ ☆ K-SHAP: Policy Clustering Algorithm for Anonymous State-Action Pairs ICML 2023
Learning agent behaviors from observational data has shown to improve our understanding of their decision-making processes, advancing our ability to explain their interactions with the environment and other agents. While multiple learning techniques have been proposed in the literature, there is one particular setting that has not been explored yet: multi agent systems where agent identities remain anonymous. For instance, in financial markets labeled data that identifies market participant strategies is typically proprietary, and only the anonymous state-action pairs that result from the interaction of multiple market participants are publicly available. As a result, sequences of agent actions are not observable, restricting the applicability of existing work. In this paper, we propose a Policy Clustering algorithm, called K-SHAP, that learns to group anonymous state-action pairs according to the agent policies. We frame the problem as an Imitation Learning (IL) task, and we learn a world-policy able to mimic all the agent behaviors upon different environmental states. We leverage the world-policy to explain each anonymous observation through an additive feature attribution method called SHAP (SHapley Additive exPlanations). Finally, by clustering the explanations we show that we are able to identify different agent policies and group observations accordingly. We evaluate our approach on simulated synthetic market data and a real-world financial dataset. We show that our proposal significantly and consistently outperforms the existing methods, identifying different agent strategies.
comment: ICML 2023
♻ ☆ On Sampling with Approximate Transport Maps
Transport maps can ease the sampling of distributions with non-trivial geometries by transforming them into distributions that are easier to handle. The potential of this approach has risen with the development of Normalizing Flows (NF) which are maps parameterized with deep neural networks trained to push a reference distribution towards a target. NF-enhanced samplers recently proposed blend (Markov chain) Monte Carlo methods with either (i) proposal draws from the flow or (ii) a flow-based reparametrization. In both cases, the quality of the learned transport conditions performance. The present work clarifies for the first time the relative strengths and weaknesses of these two approaches. Our study concludes that multimodal targets can be reliably handled with flow-based proposals up to moderately high dimensions. In contrast, methods relying on reparametrization struggle with multimodality but are more robust otherwise in high-dimensional settings and under poor training. To further illustrate the influence of target-proposal adequacy, we also derive a new quantitative bound for the mixing time of the Independent Metropolis-Hastings sampler.
♻ ☆ Can Bad Teaching Induce Forgetting? Unlearning in Deep Networks using an Incompetent Teacher AAAI 2023
Machine unlearning has become an important area of research due to an increasing need for machine learning (ML) applications to comply with the emerging data privacy regulations. It facilitates the provision for removal of certain set or class of data from an already trained ML model without requiring retraining from scratch. Recently, several efforts have been put in to make unlearning to be effective and efficient. We propose a novel machine unlearning method by exploring the utility of competent and incompetent teachers in a student-teacher framework to induce forgetfulness. The knowledge from the competent and incompetent teachers is selectively transferred to the student to obtain a model that doesn't contain any information about the forget data. We experimentally show that this method generalizes well, is fast and effective. Furthermore, we introduce the zero retrain forgetting (ZRF) metric to evaluate any unlearning method. Unlike the existing unlearning metrics, the ZRF score does not depend on the availability of the expensive retrained model. This makes it useful for analysis of the unlearned model after deployment as well. We present results of experiments conducted for random subset forgetting and class forgetting on various deep networks and across different application domains.~Source code is at: https://github.com/vikram2000b/bad-teaching-unlearning
comment: Accepted in AAAI 2023
♻ ☆ Graph Neural Networks can Recover the Hidden Features Solely from the Graph Structure ICML 2023
Graph Neural Networks (GNNs) are popular models for graph learning problems. GNNs show strong empirical performance in many practical tasks. However, the theoretical properties have not been completely elucidated. In this paper, we investigate whether GNNs can exploit the graph structure from the perspective of the expressive power of GNNs. In our analysis, we consider graph generation processes that are controlled by hidden (or latent) node features, which contain all information about the graph structure. A typical example of this framework is kNN graphs constructed from the hidden features. In our main results, we show that GNNs can recover the hidden node features from the input graph alone, even when all node features, including the hidden features themselves and any indirect hints, are unavailable. GNNs can further use the recovered node features for downstream tasks. These results show that GNNs can fully exploit the graph structure by themselves, and in effect, GNNs can use both the hidden and explicit node features for downstream tasks. In the experiments, we confirm the validity of our results by showing that GNNs can accurately recover the hidden features using a GNN architecture built based on our theoretical analysis.
comment: ICML 2023
♻ ☆ Yggdrasil Decision Forests: A Fast and Extensible Decision Forests Library
Yggdrasil Decision Forests is a library for the training, serving and interpretation of decision forest models, targeted both at research and production work, implemented in C++, and available in C++, command line interface, Python (under the name TensorFlow Decision Forests), JavaScript, Go, and Google Sheets (under the name Simple ML for Sheets). The library has been developed organically since 2018 following a set of four design principles applicable to machine learning libraries and frameworks: simplicity of use, safety of use, modularity and high-level abstraction, and integration with other machine learning libraries. In this paper, we describe those principles in detail and present how they have been used to guide the design of the library. We then showcase the use of our library on a set of classical machine learning problems. Finally, we report a benchmark comparing our library to related solutions.
♻ ☆ RoMFAC: A robust mean-field actor-critic reinforcement learning against adversarial perturbations on states
Multi-agent deep reinforcement learning makes optimal decisions dependent on system states observed by agents, but any uncertainty on the observations may mislead agents to take wrong actions. The Mean-Field Actor-Critic reinforcement learning (MFAC) is well-known in the multi-agent field since it can effectively handle a scalability problem. However, it is sensitive to state perturbations that can significantly degrade the team rewards. This work proposes a Robust Mean-field Actor-Critic reinforcement learning (RoMFAC) that has two innovations: 1) a new objective function of training actors, composed of a \emph{policy gradient function} that is related to the expected cumulative discount reward on sampled clean states and an \emph{action loss function} that represents the difference between actions taken on clean and adversarial states; and 2) a repetitive regularization of the action loss, ensuring the trained actors to obtain excellent performance. Furthermore, this work proposes a game model named a State-Adversarial Stochastic Game (SASG). Despite the Nash equilibrium of SASG may not exist, adversarial perturbations to states in the RoMFAC are proven to be defensible based on SASG. Experimental results show that RoMFAC is robust against adversarial perturbations while maintaining its competitive performance in environments without perturbations.
♻ ☆ Regression with Sensor Data Containing Incomplete Observations ICML 2023
This paper addresses a regression problem in which output label values are the results of sensing the magnitude of a phenomenon. A low value of such labels can mean either that the actual magnitude of the phenomenon was low or that the sensor made an incomplete observation. This leads to a bias toward lower values in labels and the resultant learning because labels may have lower values due to incomplete observations, even if the actual magnitude of the phenomenon was high. Moreover, because an incomplete observation does not provide any tags indicating incompleteness, we cannot eliminate or impute them. To address this issue, we propose a learning algorithm that explicitly models incomplete observations corrupted with an asymmetric noise that always has a negative value. We show that our algorithm is unbiased as if it were learned from uncorrupted data that does not involve incomplete observations. We demonstrate the advantages of our algorithm through numerical experiments.
comment: Accepted to ICML 2023
♻ ☆ Deep Regression Unlearning ICML 2023
With the introduction of data protection and privacy regulations, it has become crucial to remove the lineage of data on demand from a machine learning (ML) model. In the last few years, there have been notable developments in machine unlearning to remove the information of certain training data efficiently and effectively from ML models. In this work, we explore unlearning for the regression problem, particularly in deep learning models. Unlearning in classification and simple linear regression has been considerably investigated. However, unlearning in deep regression models largely remains an untouched problem till now. In this work, we introduce deep regression unlearning methods that generalize well and are robust to privacy attacks. We propose the Blindspot unlearning method which uses a novel weight optimization process. A randomly initialized model, partially exposed to the retain samples and a copy of the original model are used together to selectively imprint knowledge about the data that we wish to keep and scrub off the information of the data we wish to forget. We also propose a Gaussian fine tuning method for regression unlearning. The existing unlearning metrics for classification are not directly applicable to regression unlearning. Therefore, we adapt these metrics for the regression setting. We conduct regression unlearning experiments for computer vision, natural language processing and forecasting applications. Our methods show excellent performance for all these datasets across all the metrics. Source code: https://github.com/ayu987/deep-regression-unlearning
comment: Accepted in ICML 2023
♻ ☆ Generalizing Neural Wave Functions ICML 2023
Recent neural network-based wave functions have achieved state-of-the-art accuracies in modeling ab-initio ground-state potential energy surface. However, these networks can only solve different spatial arrangements of the same set of atoms. To overcome this limitation, we present Graph-learned orbital embeddings (Globe), a neural network-based reparametrization method that can adapt neural wave functions to different molecules. Globe learns representations of local electronic structures that generalize across molecules via spatial message passing by connecting molecular orbitals to covalent bonds. Further, we propose a size-consistent wave function Ansatz, the Molecular orbital network (Moon), tailored to jointly solve Schr\"odinger equations of different molecules. In our experiments, we find Moon converging in 4.5 times fewer steps to similar accuracy as previous methods or to lower energies given the same time. Further, our analysis shows that Moon's energy estimate scales additively with increased system sizes, unlike previous work where we observe divergence. In both computational chemistry and machine learning, we are the first to demonstrate that a single wave function can solve the Schr\"odinger equation of molecules with different atoms jointly.
comment: Published at the 40th International Conference on Machine Learning (ICML 2023)
♻ ☆ Efficient Online Reinforcement Learning with Offline Data ICML 2023
Sample efficiency and exploration remain major challenges in online reinforcement learning (RL). A powerful approach that can be applied to address these issues is the inclusion of offline data, such as prior trajectories from a human expert or a sub-optimal exploration policy. Previous methods have relied on extensive modifications and additional complexity to ensure the effective use of this data. Instead, we ask: can we simply apply existing off-policy methods to leverage offline data when learning online? In this work, we demonstrate that the answer is yes; however, a set of minimal but important changes to existing off-policy RL algorithms are required to achieve reliable performance. We extensively ablate these design choices, demonstrating the key factors that most affect performance, and arrive at a set of recommendations that practitioners can readily apply, whether their data comprise a small number of expert demonstrations or large volumes of sub-optimal trajectories. We see that correct application of these simple recommendations can provide a $\mathbf{2.5\times}$ improvement over existing approaches across a diverse set of competitive benchmarks, with no additional computational overhead. We have released our code at https://github.com/ikostrikov/rlpd.
comment: Short Presentation at ICML 2023; to reproduce our results and use our codebase, see https://github.com/ikostrikov/rlpd
♻ ☆ Variational Open-Domain Question Answering ICML 2023
Retrieval-augmented models have proven to be effective in natural language processing tasks, yet there remains a lack of research on their optimization using variational inference. We introduce the Variational Open-Domain (VOD) framework for end-to-end training and evaluation of retrieval-augmented models, focusing on open-domain question answering and language modelling. The VOD objective, a self-normalized estimate of the R\'enyi variational bound, approximates the task marginal likelihood and is evaluated under samples drawn from an auxiliary sampling distribution (cached retriever and/or approximate posterior). It remains tractable, even for retriever distributions defined on large corpora. We demonstrate VOD's versatility by training reader-retriever BERT-sized models on multiple-choice medical exam questions. On the MedMCQA dataset, we outperform the domain-tuned Med-PaLM by +5.3% despite using 2.500$\times$ fewer parameters. Our retrieval-augmented BioLinkBERT model scored 62.9% on the MedMCQA and 55.0% on the MedQA-USMLE. Last, we show the effectiveness of our learned retriever component in the context of medical semantic search.
comment: 28 pages, 5 figures. Accepted at ICML 2023
♻ ☆ The Curse of Recursion: Training on Generated Data Makes Models Forget
Stable Diffusion revolutionised image creation from descriptive text. GPT-2, GPT-3(.5) and GPT-4 demonstrated astonishing performance across a variety of language tasks. ChatGPT introduced such language models to the general public. It is now clear that large language models (LLMs) are here to stay, and will bring about drastic change in the whole ecosystem of online text and images. In this paper we consider what the future might hold. What will happen to GPT-{n} once LLMs contribute much of the language found online? We find that use of model-generated content in training causes irreversible defects in the resulting models, where tails of the original content distribution disappear. We refer to this effect as Model Collapse and show that it can occur in Variational Autoencoders, Gaussian Mixture Models and LLMs. We build theoretical intuition behind the phenomenon and portray its ubiquity amongst all learned generative models. We demonstrate that it has to be taken seriously if we are to sustain the benefits of training from large-scale data scraped from the web. Indeed, the value of data collected about genuine human interactions with systems will be increasingly valuable in the presence of content generated by LLMs in data crawled from the Internet.
♻ ☆ Understanding convolution on graphs via energies
Graph Neural Networks (GNNs) typically operate by message-passing, where the state of a node is updated based on the information received from its neighbours. Most message-passing models act as graph convolutions, where features are mixed by a shared, linear transformation before being propagated over the edges. On node-classification tasks, graph convolutions have been shown to suffer from two limitations: poor performance on heterophilic graphs, and over-smoothing. It is common belief that both phenomena occur because such models behave as low-pass filters, meaning that the Dirichlet energy of the features decreases along the layers incurring a smoothing effect that ultimately makes features no longer distinguishable. In this work, we rigorously prove that simple graph-convolutional models can actually enhance high frequencies and even lead to an asymptotic behaviour we refer to as over-sharpening, opposite to over-smoothing. We do so by showing that linear graph convolutions with symmetric weights minimize a multi-particle energy that generalizes the Dirichlet energy; in this setting, the weight matrices induce edge-wise attraction (repulsion) through their positive (negative) eigenvalues, thereby controlling whether the features are being smoothed or sharpened. We also extend the analysis to non-linear GNNs, and demonstrate that some existing time-continuous GNNs are instead always dominated by the low frequencies. Finally, we validate our theoretical findings through ablations and real-world experiments.
comment: First two authors equal contribution; 34 pages
♻ ☆ Modeling Dynamic Environments with Scene Graph Memory
Embodied AI agents that search for objects in large environments such as households often need to make efficient decisions by predicting object locations based on partial information. We pose this as a new type of link prediction problem: link prediction on partially observable dynamic graphs. Our graph is a representation of a scene in which rooms and objects are nodes, and their relationships are encoded in the edges; only parts of the changing graph are known to the agent at each timestep. This partial observability poses a challenge to existing link prediction approaches, which we address. We propose a novel state representation -- Scene Graph Memory (SGM) -- with captures the agent's accumulated set of observations, as well as a neural net architecture called a Node Edge Predictor (NEP) that extracts information from the SGM to search efficiently. We evaluate our method in the Dynamic House Simulator, a new benchmark that creates diverse dynamic graphs following the semantic patterns typically seen at homes, and show that NEP can be trained to predict the locations of objects in a variety of environments with diverse object movement dynamics, outperforming baselines both in terms of new scene adaptability and overall accuracy. The codebase and more can be found at https://www.scenegraphmemory.com.
♻ ☆ Is My Prediction Arbitrary? Measuring Self-Consistency in Fair Classification
Variance in predictions across different trained models is a significant, under-explored source of error in fair classification. Empirically, the variance on some instances is so large that decisions can be effectively arbitrary. To study this problem, we perform a large-scale empirical study and make four overarching contributions: We 1) Define a metric called self-consistency, derived from variance, which we use as a proxy for measuring and reducing arbitrariness; 2) Develop an ensembling algorithm that abstains from classification when a prediction would be arbitrary; 3) Conduct the largest to-date empirical study of the role of variance (vis-a-vis self-consistency and arbitrariness) in fair classification; and, 4) Release a toolkit that makes the US Home Mortgage Disclosure Act (HMDA) datasets easily usable for future research. Altogether, our empirical results reveal shocking insights about reproducibility. Most fairness classification benchmarks are close-to-fair when taking into account the amount of arbitrariness present in predictions. Subgroup error rates are similar before we even try to apply common fairness interventions
♻ ☆ UPop: Unified and Progressive Pruning for Compressing Vision-Language Transformers ICML 2023
Real-world data contains a vast amount of multimodal information, among which vision and language are the two most representative modalities. Moreover, increasingly heavier models, \textit{e}.\textit{g}., Transformers, have attracted the attention of researchers to model compression. However, how to compress multimodal models, especially vison-language Transformers, is still under-explored. This paper proposes the \textbf{U}nified and \textbf{P}r\textbf{o}gressive \textbf{P}runing (\textbf{\emph{UPop}}) as a universal vison-language Transformer compression framework, which incorporates 1) unifiedly searching multimodal subnets in a continuous optimization space from the original model, which enables automatic assignment of pruning ratios among compressible modalities and structures; 2) progressively searching and retraining the subnet, which maintains convergence between the search and retrain to attain higher compression ratios. Experiments on various tasks, datasets, and model architectures demonstrate the effectiveness and versatility of the proposed UPop framework. The code is available at https://github.com/sdc17/UPop.
comment: ICML 2023
♻ ☆ Friendly Neighbors: Contextualized Sequence-to-Sequence Link Prediction
We propose KGT5-context, a simple sequence-to-sequence model for link prediction (LP) in knowledge graphs (KG). Our work expands on KGT5, a recent LP model that exploits textual features of the KG, has small model size, and is scalable. To reach good predictive performance, however, KGT5 relies on an ensemble with a knowledge graph embedding model, which itself is excessively large and costly to use. In this short paper, we show empirically that adding contextual information - i.e., information about the direct neighborhood of the query entity - alleviates the need for a separate KGE model to obtain good performance. The resulting KGT5-context model is simple, reduces model size significantly, and obtains state-of-the-art performance in our experimental study.
comment: 7 pages, 2 figures
♻ ☆ How Powerful are Shallow Neural Networks with Bandlimited Random Weights? ICML 2023
We investigate the expressive power of depth-2 bandlimited random neural networks. A random net is a neural network where the hidden layer parameters are frozen with random assignment, and only the output layer parameters are trained by loss minimization. Using random weights for a hidden layer is an effective method to avoid non-convex optimization in standard gradient descent learning. It has also been adopted in recent deep learning theories. Despite the well-known fact that a neural network is a universal approximator, in this study, we mathematically show that when hidden parameters are distributed in a bounded domain, the network may not achieve zero approximation error. In particular, we derive a new nontrivial approximation error lower bound. The proof utilizes the technique of ridgelet analysis, a harmonic analysis method designed for neural networks. This method is inspired by fundamental principles in classical signal processing, specifically the idea that signals with limited bandwidth may not always be able to perfectly recreate the original signal. We corroborate our theoretical results with various simulation studies, and generally, two main take-home messages are offered: (i) Not any distribution for selecting random weights is feasible to build a universal approximator; (ii) A suitable assignment of random weights exists but to some degree is associated with the complexity of the target function.
comment: Published as a conference paper at ICML 2023
♻ ☆ CAP: Correlation-Aware Pruning for Highly-Accurate Sparse Vision Models
Driven by significant improvements in architectural design and training pipelines, computer vision has recently experienced dramatic progress in terms of accuracy on classic benchmarks such as ImageNet. These highly-accurate models are challenging to deploy, as they appear harder to compress using standard techniques such as pruning. We address this issue by introducing the Correlation Aware Pruner (CAP), a new unstructured pruning framework which significantly pushes the compressibility limits for state-of-the-art architectures. Our method is based on two technical advancements: a new theoretically-justified pruner, which can handle complex weight correlations accurately and efficiently during the pruning process itself, and an efficient finetuning procedure for post-compression recovery. We validate our approach via extensive experiments on several modern vision models such as Vision Transformers (ViT), modern CNNs, and ViT-CNN hybrids, showing for the first time that these can be pruned to high sparsity levels (e.g. $\geq 75$%) with low impact on accuracy ($\leq 1$% relative drop). Our approach is also compatible with structured pruning and quantization, and can lead to practical speedups of 1.5 to 2.4x without accuracy loss. To further showcase CAP's accuracy and scalability, we use it to show for the first time that extremely-accurate large vision models, trained via self-supervised techniques, can also be pruned to moderate sparsities, with negligible accuracy loss.
♻ ☆ Dink-Net: Neural Clustering on Large Graphs
Deep graph clustering, which aims to group the nodes of a graph into disjoint clusters with deep neural networks, has achieved promising progress in recent years. However, the existing methods fail to scale to the large graph with million nodes. To solve this problem, a scalable deep graph clustering method (Dink-Net) is proposed with the idea of dilation and shrink. Firstly, by discriminating nodes, whether being corrupted by augmentations, representations are learned in a self-supervised manner. Meanwhile, the cluster centres are initialized as learnable neural parameters. Subsequently, the clustering distribution is optimized by minimizing the proposed cluster dilation loss and cluster shrink loss in an adversarial manner. By these settings, we unify the two-step clustering, i.e., representation learning and clustering optimization, into an end-to-end framework, guiding the network to learn clustering-friendly features. Besides, Dink-Net scales well to large graphs since the designed loss functions adopt the mini-batch data to optimize the clustering distribution even without performance drops. Both experimental results and theoretical analyses demonstrate the superiority of our method. Compared to the runner-up, Dink-Net achieves 9.62% NMI improvement on the ogbn-papers100M dataset with 111 million nodes and 1.6 billion edges. The source code is released at https://github.com/yueliu1999/Dink-Net. Besides, a collection (papers, codes, and datasets) of deep graph clustering is shared at https://github.com/yueliu1999/Awesome-Deep-Graph-Clustering.
comment: 18 pages, 5 figures
Multimedia 5
☆ Off-By-One Implementation Error in J-UNIWARD
J-UNIWARD is a popular steganography method for hiding secret messages in JPEG cover images. As a content-adaptive method, J-UNIWARD aims to embed into textured image regions where changes are difficult to detect. To this end, J-UNIWARD first assigns to each DCT coefficient an embedding cost calculated based on the image's Wavelet residual, and then uses a coding method that minimizes the cost while embedding the desired payload. Changing one DCT coefficient affects a 23x23 window of Wavelet coefficients. To speed up the costmap computation, the original implementation pre-computes the Wavelet residual and then considers per changed DCT coefficient a 23x23 window of the Wavelet residual. However, the implementation accesses a window accidentally shifted by one pixel to the bottom right. In this report, we evaluate the effect of this off-by-one error on the resulting costmaps. Some image blocks are over-priced while other image blocks are under-priced, but the difference is relatively small. The off-by-one error seems to make little difference for learning-based steganalysis.
☆ Unveiling Cross Modality Bias in Visual Question Answering: A Causal View with Possible Worlds VQA
To increase the generalization capability of VQA systems, many recent studies have tried to de-bias spurious language or vision associations that shortcut the question or image to the answer. Despite these efforts, the literature fails to address the confounding effect of vision and language simultaneously. As a result, when they reduce bias learned from one modality, they usually increase bias from the other. In this paper, we first model a confounding effect that causes language and vision bias simultaneously, then propose a counterfactual inference to remove the influence of this effect. The model trained in this strategy can concurrently and efficiently reduce vision and language bias. To the best of our knowledge, this is the first work to reduce biases resulting from confounding effects of vision and language in VQA, leveraging causal explain-away relations. We accompany our method with an explain-away strategy, pushing the accuracy of the questions with numerical answers results compared to existing methods that have been an open problem. The proposed method outperforms the state-of-the-art methods in VQA-CP v2 datasets.
comment: 22 pages
☆ MuseCoco: Generating Symbolic Music from Text
Generating music from text descriptions is a user-friendly mode since the text is a relatively easy interface for user engagement. While some approaches utilize texts to control music audio generation, editing musical elements in generated audio is challenging for users. In contrast, symbolic music offers ease of editing, making it more accessible for users to manipulate specific musical elements. In this paper, we propose MuseCoco, which generates symbolic music from text descriptions with musical attributes as the bridge to break down the task into text-to-attribute understanding and attribute-to-music generation stages. MuseCoCo stands for Music Composition Copilot that empowers musicians to generate music directly from given text descriptions, offering a significant improvement in efficiency compared to creating music entirely from scratch. The system has two main advantages: Firstly, it is data efficient. In the attribute-to-music generation stage, the attributes can be directly extracted from music sequences, making the model training self-supervised. In the text-to-attribute understanding stage, the text is synthesized and refined by ChatGPT based on the defined attribute templates. Secondly, the system can achieve precise control with specific attributes in text descriptions and offers multiple control options through attribute-conditioned or text-conditioned approaches. MuseCoco outperforms baseline systems in terms of musicality, controllability, and overall score by at least 1.27, 1.08, and 1.32 respectively. Besides, there is a notable enhancement of about 20% in objective control accuracy. In addition, we have developed a robust large-scale model with 1.2 billion parameters, showcasing exceptional controllability and musicality.
♻ ☆ Multi-task Paired Masking with Alignment Modeling for Medical Vision-Language Pre-training
In recent years, the growing demand for medical imaging diagnosis has placed a significant burden on radiologists. As a solution, Medical Vision-Language Pre-training (Med-VLP) methods have been proposed to learn universal representations from medical images and reports, benefiting downstream tasks without requiring fine-grained annotations. However, existing methods have overlooked the importance of cross-modal alignment in joint image-text reconstruction, resulting in insufficient cross-modal interaction. To address this limitation, we propose a unified Med-VLP framework based on Multi-task Paired Masking with Alignment (MPMA) to integrate the cross-modal alignment task into the joint image-text reconstruction framework to achieve more comprehensive cross-modal interaction, while a Global and Local Alignment (GLA) module is designed to assist self-supervised paradigm in obtaining semantic representations with rich domain knowledge. Furthermore, we introduce a Memory-Augmented Cross-Modal Fusion (MA-CMF) module to fully integrate visual information to assist report reconstruction and fuse the multi-modal representations adequately. Experimental results demonstrate that the proposed unified approach outperforms previous methods in all downstream tasks, including uni-modal, cross-modal, and multi-modal tasks.
♻ ☆ ImageBind: One Embedding Space To Bind Them All CVPR 2023
We present ImageBind, an approach to learn a joint embedding across six different modalities - images, text, audio, depth, thermal, and IMU data. We show that all combinations of paired data are not necessary to train such a joint embedding, and only image-paired data is sufficient to bind the modalities together. ImageBind can leverage recent large scale vision-language models, and extends their zero-shot capabilities to new modalities just by using their natural pairing with images. It enables novel emergent applications 'out-of-the-box' including cross-modal retrieval, composing modalities with arithmetic, cross-modal detection and generation. The emergent capabilities improve with the strength of the image encoder and we set a new state-of-the-art on emergent zero-shot recognition tasks across modalities, outperforming specialist supervised models. Finally, we show strong few-shot recognition results outperforming prior work, and that ImageBind serves as a new way to evaluate vision models for visual and non-visual tasks.
comment: CVPR 2023 (Highlighted Paper). Website: https://imagebind.metademolab.com/ Code/Models: https://github.com/facebookresearch/ImageBind